{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "95507e34-4f85-4116-b8c7-45c1cdacfaf1",
   "metadata": {},
   "source": [
    "# Exploring three time series classifiers from the sktime library towards short-term (~ one hour) predictions of SEP events."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3d386a64-1802-4a99-b4e5-5653c15678e3",
   "metadata": {},
   "source": [
    "## Author(s)\n",
    "- Author1: Sumanth Rotti; Email: srotti@gsu.edu; ORCID: 0000-0003-1080-3424\n",
    "- Author2: Berkay Aydin; Email: baydin2@gsu.edu; ORCID: 0000-0002-9799-9265\n",
    "- Author3: Petrus C. Martens; Email: pmartens@gsu.edu; ORCID: 0000-0001-8078-6856\n",
    "- Affiliation: Georgia State University"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "7f13fb4a-f4f0-43a3-b226-4ca9eba9d14b",
   "metadata": {},
   "outputs": [],
   "source": [
    "#importing libraries\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "from numpy import argmax, arange\n",
    "from time import time\n",
    "import glob\n",
    "from typing import List, Union"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "c2285b07-8080-4b1e-8113-f1f837dedcb5",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "<frozen importlib._bootstrap>:241: RuntimeWarning: pyarrow.lib.IpcReadOptions size changed, may indicate binary incompatibility. Expected 96 from C header, got 104 from PyObject\n"
     ]
    }
   ],
   "source": [
    "#importing libraries\n",
    "import sktime\n",
    "import sktime.datatypes\n",
    "from sktime.datatypes import check_is_mtype, check_raise"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "3dfefaa9-e65f-4f25-9239-265c0f435b9e",
   "metadata": {},
   "outputs": [],
   "source": [
    "#importing libraries\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sktime.classification.compose import ColumnEnsembleClassifier\n",
    "from sktime.classification.interval_based import SupervisedTimeSeriesForest\n",
    "from sktime.classification.feature_based import SummaryClassifier\n",
    "from sktime.classification.distance_based import KNeighborsTimeSeriesClassifier\n",
    "from sktime.classification.hybrid import HIVECOTEV2\n",
    "from sktime.classification.kernel_based import RocketClassifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "2b69b443",
   "metadata": {},
   "outputs": [],
   "source": [
    "#importing libraries\n",
    "from sklearn import metrics\n",
    "from sklearn.preprocessing import binarize\n",
    "from sklearn.metrics import confusion_matrix, ConfusionMatrixDisplay, make_scorer, roc_auc_score, roc_curve, DetCurveDisplay\n",
    "from sklearn.metrics import accuracy_score, precision_score, recall_score, f1_score, matthews_corrcoef\n",
    "from sklearn.metrics import (brier_score_loss, log_loss, recall_score)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "b2ebb0fa",
   "metadata": {},
   "outputs": [],
   "source": [
    "#importing plotting libraries\n",
    "import matplotlib.pyplot as plt\n",
    "from matplotlib import pyplot\n",
    "import seaborn as sns\n",
    "from sklearn_evaluation import plot"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "78132cae",
   "metadata": {},
   "outputs": [],
   "source": [
    "import warnings\n",
    "warnings.simplefilter(action='ignore', category=FutureWarning)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "361ad498",
   "metadata": {},
   "outputs": [],
   "source": [
    "plt.rcParams.update({'font.size': 13})"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6cfb335c-0708-4cd6-8fe3-04b018fb4b22",
   "metadata": {},
   "source": [
    "# Loading the data set"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1d9c9b51-114f-4cf0-aa47-d0c7020ee693",
   "metadata": {},
   "source": [
    "- We will use the 433 (244+189) samples in the GSEP data set and an additional 2,460 negative samples introducing a natural class imbalance between SEPs and non-SEPs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "512fe2f8",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2893\n"
     ]
    }
   ],
   "source": [
    "# Reading .csv files\n",
    "extension = 'csv'\n",
    "csv_filepath_list  = sorted([i for i in glob.glob(r'./data/*.{}'.format(extension))])\n",
    "print(len(csv_filepath_list))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "7b64986f",
   "metadata": {},
   "outputs": [],
   "source": [
    "# List of column headers to read from the csv files\n",
    "features = ['p3_flux_ic','p5_flux_ic','p7_flux_ic','long']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "17d2bd63",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2893"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Read and convert each .csv file individually\n",
    "csv_list = []\n",
    "for csv_index, csv_path_i in enumerate(csv_filepath_list):\n",
    "    #Considering 11 hr of observation minus 60 minutes before the SEP event onset\n",
    "    csv_df_i = pd.read_csv(csv_path_i, sep=',', header=0, nrows = 660, usecols = features)\n",
    "    #append sktime DataFrames to a list\n",
    "    csv_list.append(csv_df_i)\n",
    "len(csv_list)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "a4631494",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(True,\n",
       " None,\n",
       " {'is_univariate': False,\n",
       "  'is_equally_spaced': True,\n",
       "  'is_equal_length': True,\n",
       "  'is_empty': False,\n",
       "  'has_nans': False,\n",
       "  'is_one_series': False,\n",
       "  'n_panels': 1,\n",
       "  'is_one_panel': True,\n",
       "  'n_instances': 2893,\n",
       "  'n_features': 4,\n",
       "  'feature_names': ['p3_flux_ic', 'p5_flux_ic', 'p7_flux_ic', 'long'],\n",
       "  'mtype': 'df-list',\n",
       "  'scitype': 'Panel'})"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Checking for sktime data format compatibility\n",
    "check_is_mtype(csv_list, mtype=\"df-list\", return_metadata=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "2f6ed72d",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Loading  the class label file \n",
    "tgt = pd.read_csv(\"labels.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "10d8e73b",
   "metadata": {},
   "outputs": [],
   "source": [
    "#renaming the label column to 'target'\n",
    "tgt.rename(columns = {'Label':'target'}, inplace = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "0fa8c2f5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>File</th>\n",
       "      <th>target</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1986-01-15_05-54.csv</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1986-01-15_19-57.csv</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1986-01-16_15-16.csv</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                   File  target\n",
       "0  1986-01-15_05-54.csv       0\n",
       "1  1986-01-15_19-57.csv       0\n",
       "2  1986-01-16_15-16.csv       0"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tgt.head(3)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9267d18c",
   "metadata": {},
   "source": [
    "# Splitting the data set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "db8ddc38-0ee1-479f-ad51-6c628698e2c6",
   "metadata": {},
   "outputs": [],
   "source": [
    "# training set\n",
    "X_train=csv_list[0:998]\n",
    "y_train=tgt['target'][0:998]\n",
    "# validation set\n",
    "X_valid=csv_list[998:1972]\n",
    "y_valid=tgt['target'][998:1972]\n",
    "# test set\n",
    "X_test=csv_list[1972:]\n",
    "y_test=tgt['target'][1972:]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1c5999fc",
   "metadata": {},
   "source": [
    "## Check the size of each data splits"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "2f860381",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Length of training set:  998\n",
      "Length of validation set:  974\n",
      "Length of test set:  921\n"
     ]
    }
   ],
   "source": [
    "print(\"Length of training set: \", len(X_train))\n",
    "print(\"Length of validation set: \", len(X_valid))\n",
    "print(\"Length of test set: \", len(X_test))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c73ad2e7-8f06-4c21-bca3-b4484fef389b",
   "metadata": {},
   "source": [
    "## Check the splitting size with respect to target"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "efa3e6ad-8813-4eea-ab66-3fdd3dac3411",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[  0 918]\n",
      " [  1  80]]\n"
     ]
    }
   ],
   "source": [
    "unique, counts = np.unique(y_train, return_counts=True)\n",
    "print(np.asarray((unique, counts)).T) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "4646ccf8-a99e-4696-83d3-a3111c75417f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[  0 894]\n",
      " [  1  80]]\n"
     ]
    }
   ],
   "source": [
    "unique, counts = np.unique(y_valid, return_counts=True)\n",
    "print(np.asarray((unique, counts)).T) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "53f067b9-cfd7-464b-a4c7-d2d81d2f5c66",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[  0 837]\n",
      " [  1  84]]\n"
     ]
    }
   ],
   "source": [
    "unique, counts = np.unique(y_test, return_counts=True)\n",
    "print(np.asarray((unique, counts)).T) "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e809fa3f",
   "metadata": {},
   "source": [
    "# Defining custom metrics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "e35d05e2-58fd-4db5-8feb-7b2c52cbe684",
   "metadata": {},
   "outputs": [],
   "source": [
    "def tss(y, pred):\n",
    "    '''\n",
    "    Function to estimate true skill statistics (TSS) \n",
    "    \n",
    "    Inputs:\n",
    "    y - True labels\n",
    "    pred - Predicted labels\n",
    "    \n",
    "    Returns:\n",
    "    The TSS score for given input data\n",
    "    '''\n",
    "    TN, FP, FN, TP = confusion_matrix(y, pred).ravel()\n",
    "    #\n",
    "    x = TP * TN\n",
    "    y = FP * FN\n",
    "    A = TP + FN\n",
    "    B = FN + TN\n",
    "    C = TP + FP\n",
    "    D = FP + TN\n",
    "    #\n",
    "    tot = TP + FP + FN + TN\n",
    "    ref = ((A * C) / tot)\n",
    "    #\n",
    "    tss = round(((x - y) / (A * D)), 3)\n",
    "    #\n",
    "    return tss"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "bbc5e036-aa9a-4e21-b59d-13ca42de619a",
   "metadata": {},
   "outputs": [],
   "source": [
    "def hss(y, pred):\n",
    "    '''\n",
    "    Function to estimate Heidke skill statistics (TSS) \n",
    "    \n",
    "    Inputs:\n",
    "    y - True labels\n",
    "    pred - Predicted labels\n",
    "    \n",
    "    Returns:\n",
    "    The HSS score for given input data\n",
    "    '''\n",
    "    TN, FP, FN, TP = confusion_matrix(y, pred).ravel()\n",
    "    #\n",
    "    x = TP * TN\n",
    "    y = FP * FN\n",
    "    A = TP + FN\n",
    "    B = FN + TN\n",
    "    C = TP + FP\n",
    "    D = FP + TN\n",
    "    #\n",
    "    tot = TP + FP + FN + TN\n",
    "    ref = ((A * C) / tot)\n",
    "    #\n",
    "    hss = round((2 * (x-y)/((A * B) + (C * D))), 3)\n",
    "    #\n",
    "    return hss"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "f6a3df33-9be3-4cc8-a0b8-39d006133b31",
   "metadata": {},
   "outputs": [],
   "source": [
    "def gss(y, pred):\n",
    "    '''\n",
    "    Function to estimate Gilbert skill statistics (TSS) \n",
    "    \n",
    "    Inputs:\n",
    "    y - True labels\n",
    "    pred - Predicted labels\n",
    "    \n",
    "    Returns:\n",
    "    The GSS score for given input data\n",
    "    '''\n",
    "    TN, FP, FN, TP = confusion_matrix(y, pred).ravel()\n",
    "    #\n",
    "    x = TP * TN\n",
    "    y = FP * FN\n",
    "    A = TP + FN\n",
    "    B = FN + TN\n",
    "    C = TP + FP\n",
    "    D = FP + TN\n",
    "    #\n",
    "    tot = TP + FP + FN + TN\n",
    "    ref = ((A * C) / tot)\n",
    "    #\n",
    "    gss = round(((TP - ref) / (TP + FP + FN - ref)), 3)\n",
    "    #\n",
    "    return gss"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d4bde416-dda1-4834-ac99-ba92d5faba42",
   "metadata": {},
   "source": [
    "# Defining a function to obtain the optimal classification threshold"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "49ca6044-92e0-4630-ab63-a8a68e9535ba",
   "metadata": {},
   "outputs": [],
   "source": [
    "# apply threshold to positive probabilities to create labels\n",
    "def to_labels(pos_proba, threshold):\n",
    "    '''\n",
    "    Function to convert probabilities to labels for a specific classification threshold\n",
    "    \n",
    "    Inputs:\n",
    "    pos_proba - Predicted probability of input data point belonging to the respective class\n",
    "    threshold - Threhsold defining the boundary to separate the target class labels\n",
    "    \n",
    "    Returns:\n",
    "    Labels for the predicted probabilities \n",
    "    '''\n",
    "    return (pos_proba >= threshold).astype('int')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "1702492e-42ed-4729-af56-b163ec23f533",
   "metadata": {},
   "outputs": [],
   "source": [
    "# define thresholds as a numpy array\n",
    "thresholds_cust = arange(0, 1, 0.01)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "9cd77088",
   "metadata": {},
   "outputs": [],
   "source": [
    "rand=3"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d5b594ae-d292-4edb-8835-1e057aa57715",
   "metadata": {},
   "source": [
    "# Summary Classifier"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0c0da770-979d-40b3-a4cc-8411d644520c",
   "metadata": {},
   "source": [
    "## Training set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "1afca3d0-040f-4de7-960c-828c89500fcd",
   "metadata": {},
   "outputs": [],
   "source": [
    "ssc_train_run_start = time()\n",
    "summ = ColumnEnsembleClassifier(\n",
    "    estimators=[\n",
    "        (\"summ_0\", SummaryClassifier(estimator=RandomForestClassifier(n_estimators=200), random_state=rand), [\"p3_flux_ic\"]),\n",
    "        (\"summ_1\", SummaryClassifier(estimator=RandomForestClassifier(n_estimators=200), random_state=rand), [\"p5_flux_ic\"]),\n",
    "        (\"summ_2\", SummaryClassifier(estimator=RandomForestClassifier(n_estimators=200), random_state=rand), [\"p7_flux_ic\"]),\n",
    "        (\"summ_3\", SummaryClassifier(estimator=RandomForestClassifier(n_estimators=200), random_state=rand), [\"long\"])\n",
    "     ]\n",
    ")\n",
    "summ.fit(X_train, y_train)\n",
    "ssc_train_run_end = time()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "cf152a1a-276d-48f1-9fc7-f395761597d7",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training time of the SSC model -- 42.000000\n"
     ]
    }
   ],
   "source": [
    "# Get the model training run time (in seconds)\n",
    "ssc_train_run_time = round((ssc_train_run_end - ssc_train_run_start))\n",
    "print('Training time of the SSC model -- %f' % (ssc_train_run_time))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "052fc552-e168-4360-b950-bf11250b2f76",
   "metadata": {},
   "source": [
    "## Validation set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "4c26206b-8ac6-42ad-9a2c-9d38d37fc687",
   "metadata": {},
   "outputs": [],
   "source": [
    "# predict probabilities on the validation set\n",
    "ssc_run_start = time()\n",
    "probs_summ = summ.predict_proba(X_valid)[:,1]\n",
    "ssc_run_end = time()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "b81556a3-1861-4fd1-953d-60aeab389c93",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best Threshold = 0.156460\n"
     ]
    }
   ],
   "source": [
    "# calculate ROC curves\n",
    "fpr_summ, tpr_summ, thresholds_summ = roc_curve(y_valid, probs_summ)\n",
    "# get the best threshold using  Youden’s J statistic\n",
    "J_summ = tpr_summ - fpr_summ\n",
    "ix_summ = argmax(J_summ)\n",
    "best_thre_summ = thresholds_summ[ix_summ]\n",
    "print('Best Threshold = %f' % (best_thre_summ))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "5543f93c-94c0-4768-b92f-c1212cb0a400",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plot the roc curve for the model\n",
    "auc_summ = round(roc_auc_score(y_valid, probs_summ), 3)\n",
    "pyplot.plot([0,1], [0,1], linestyle='--', color = 'dimgrey', label='Baseline')\n",
    "pyplot.plot(fpr_summ, tpr_summ, marker='.', color = 'darkorange', markersize=5, label=\"SC, AUC=\"+str(auc_summ))\n",
    "pyplot.scatter(fpr_summ[ix_summ], tpr_summ[ix_summ], marker='*', s=70, color='dodgerblue', label='Optimal', zorder=2)\n",
    "# axis labels\n",
    "pyplot.xlabel('False Positive Rate')\n",
    "pyplot.ylabel('True Positive Rate')\n",
    "pyplot.legend()\n",
    "# displaying the title\n",
    "plt.title(\"Summary Classifier ROC curve\")\n",
    "# show the plot\n",
    "pyplot.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "7f1cbe92-8ffa-4530-8c85-77fc30eb513c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Prediction run time of SSC-- 38.000000\n"
     ]
    }
   ],
   "source": [
    "# Get the model run time\n",
    "ssc_run_time = round((ssc_run_end - ssc_run_start))\n",
    "# evaluate scores for each threshold\n",
    "f_score_summ = [f1_score(y_valid, to_labels(probs_summ, t), average='weighted') for t in thresholds_cust]\n",
    "tss_score_summ = [tss(y_valid, to_labels(probs_summ, t)) for t in thresholds_cust]\n",
    "hss_score_summ = [hss(y_valid, to_labels(probs_summ, t)) for t in thresholds_cust]\n",
    "gss_score_summ = [gss(y_valid, to_labels(probs_summ, t)) for t in thresholds_cust]\n",
    "mcc_score_summ = [matthews_corrcoef(y_valid, to_labels(probs_summ, t)) for t in thresholds_cust]\n",
    "# get best threshold\n",
    "f1_ix_summ = argmax(f_score_summ)\n",
    "tss_ix_summ = argmax(tss_score_summ)\n",
    "hss_ix_summ = argmax(hss_score_summ)\n",
    "gss_ix_summ = argmax(gss_score_summ)\n",
    "mcc_ix_summ = argmax(mcc_score_summ)\n",
    "print('Prediction run time of SSC-- %f' % (ssc_run_time))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e9cfda9e-2544-4c99-90a4-aafb73ccda0e",
   "metadata": {},
   "source": [
    "### Visualizing the variation in scores for different thresholds."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "7f971604-76ec-4ce3-99fc-89defa7dafa2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1300x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(13,5))\n",
    "pyplot.plot(thresholds_cust, np.array(tss_score_summ), marker='o', markersize=4, color='indianred',  label='TSS')\n",
    "pyplot.plot(thresholds_cust, np.array(hss_score_summ), marker='*', markersize=4, color='forestgreen', label='HSS')\n",
    "pyplot.plot(thresholds_cust, np.array(gss_score_summ), marker='s', markersize=4, color='goldenrod', label='GSS')\n",
    "pyplot.plot(thresholds_cust, np.array(mcc_score_summ), marker='v', markersize=4, color='cornflowerblue', label='MCC')\n",
    "plt.axvline(x = thresholds_cust[hss_ix_summ], color=\"maroon\", linestyle=(0, (5, 5)))\n",
    "plt.text(0.25, 0.6, r'Chosen threshold ='+str(thresholds_cust[hss_ix_summ]))\n",
    "plt.xticks(np.arange(0.0, 1.1, step=0.1))\n",
    "# axis labels\n",
    "pyplot.xlabel('Threshold')\n",
    "pyplot.ylabel('Score')\n",
    "pyplot.legend()\n",
    "# displaying the title\n",
    "plt.title(\"Summary Classifier\")\n",
    "# show the plot\n",
    "pyplot.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "860fc08e-8150-4000-9192-599725355d3a",
   "metadata": {},
   "outputs": [],
   "source": [
    "#Choosing the best classification threshold of the SSC model\n",
    "best_summ = thresholds_cust[hss_ix_summ]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "5b34c4c6-e18c-443e-927c-af632207177a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best Threshold = 0.170\n",
      "Scores -- F1_Score: 0.935 / TSS: 0.675000 / HSS: 0.597000 / GSS: 0.425000 / MCC: 0.602385\n"
     ]
    }
   ],
   "source": [
    "# evaluate the best threshold\n",
    "f1_scorex_summ = f1_score(y_valid, to_labels(probs_summ, best_summ), average='weighted')\n",
    "tss_scorex_summ = tss(y_valid, to_labels(probs_summ, best_summ))\n",
    "hss_scorex_summ = hss(y_valid, to_labels(probs_summ, best_summ))\n",
    "gss_scorex_summ = gss(y_valid, to_labels(probs_summ, best_summ))\n",
    "mcc_scorex_summ = matthews_corrcoef(y_valid, to_labels(probs_summ, best_summ))\n",
    "print('Best Threshold = %.3f' %(best_summ))\n",
    "print('Scores -- F1_Score: %.3f / TSS: %f / HSS: %f / GSS: %f / MCC: %f' % (f1_scorex_summ, tss_scorex_summ, hss_scorex_summ,\n",
    "                                                                            gss_scorex_summ, mcc_scorex_summ))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e6e8f6b4-cd00-49e8-8f16-e8f14b2bd4fc",
   "metadata": {},
   "source": [
    "### Validation set confusion matrix on the best threshold"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "2e8dc212-0d5f-4208-80fe-8f8dd7b32e17",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "y_pred_class_summ = to_labels(probs_summ, best_summ)\n",
    "color = 'white'\n",
    "summ_matrix = ConfusionMatrixDisplay.from_predictions(y_valid, y_pred_class_summ,\n",
    "                                                     cmap=plt.cm.Blues, colorbar=False)\n",
    "summ_matrix.ax_.set(title='Summary Classifier')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ae8f3e31-9a7f-480e-b355-d5be4f7dddf6",
   "metadata": {},
   "source": [
    "## Test set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "5881a27d-5751-44c5-b726-0a50961db893",
   "metadata": {},
   "outputs": [],
   "source": [
    "# predict probabilities on the test set\n",
    "probX_summ = summ.predict_proba(X_test)[:,1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "f6b42364-6ec2-42e1-a358-28fd42b88c97",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test scores -- F1_Score: 0.935 / TSS: 0.720 / HSS: 0.632 / GSS: 0.462 / MCC: 0.639\n"
     ]
    }
   ],
   "source": [
    "# evaluate on the best threshold\n",
    "f1_test_summ = (f1_score(y_test, to_labels(probX_summ, best_summ), average='weighted'))\n",
    "tss_test_summ = tss(y_test, to_labels(probX_summ, best_summ))\n",
    "hss_test_summ = hss(y_test, to_labels(probX_summ, best_summ))\n",
    "gss_test_summ = gss(y_test, to_labels(probX_summ, best_summ))\n",
    "mcc_test_summ = matthews_corrcoef(y_test, to_labels(probX_summ, best_summ))\n",
    "print('Test scores -- F1_Score: %.3f / TSS: %.3f / HSS: %.3f / GSS: %.3f / MCC: %.3f' % (f1_test_summ, tss_test_summ, \n",
    "                                                                          hss_test_summ, gss_test_summ,\n",
    "                                                                          mcc_test_summ))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fa7c90ff-06c4-4ccd-95af-9a210478ee84",
   "metadata": {},
   "source": [
    "### Test set confusion matrix on the best threshold"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "0f2cbf97-c103-46cb-b095-61c4b72364d8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "color = 'white'\n",
    "y_pred_test_summ = to_labels(probX_summ, best_summ)\n",
    "summ_matrix = ConfusionMatrixDisplay.from_predictions(y_test, y_pred_test_summ,\n",
    "                                                     cmap=plt.cm.Blues, colorbar=False)\n",
    "summ_matrix.ax_.set(title='Summary Classifier')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b15a60ad",
   "metadata": {},
   "source": [
    "# SupervisedTimeSeriesForest"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "aed8923c",
   "metadata": {},
   "source": [
    "## Training set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "fc8209c2",
   "metadata": {},
   "outputs": [],
   "source": [
    "stsf_train_run_start = time()\n",
    "stsf = ColumnEnsembleClassifier(\n",
    "    estimators=[\n",
    "        (\"stsf_0\", SupervisedTimeSeriesForest(n_estimators=200, random_state=rand), [\"p3_flux_ic\"]),\n",
    "        (\"stsf_1\", SupervisedTimeSeriesForest(n_estimators=200, random_state=rand), [\"p5_flux_ic\"]),\n",
    "        (\"stsf_2\", SupervisedTimeSeriesForest(n_estimators=200, random_state=rand), [\"p7_flux_ic\"]),\n",
    "        (\"stsf_3\", SupervisedTimeSeriesForest(n_estimators=200, random_state=rand), [\"long\"])\n",
    "     ]\n",
    ")\n",
    "stsf.fit(X_train, y_train)\n",
    "stsf_train_run_end = time()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "3fc8003f-9328-4873-a350-91c8ab13c15e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training time of the STSF model -- 676.000000\n"
     ]
    }
   ],
   "source": [
    "# Get the model training run time (in seconds)\n",
    "stsf_train_run_time = round((stsf_train_run_end - stsf_train_run_start))\n",
    "print('Training time of the STSF model -- %f' % (stsf_train_run_time))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d14b879e-1c93-44f4-81e2-3716ff081b31",
   "metadata": {},
   "source": [
    "## Validation set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "e0e8b523-d7d2-466b-b682-d5c33e0b4a30",
   "metadata": {},
   "outputs": [],
   "source": [
    "# predict probabilities on the validation set\n",
    "stsf_run_start = time()\n",
    "probs_stsf = stsf.predict_proba(X_valid)[:,1]\n",
    "stsf_run_end = time()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "64179c63-4a67-43a6-861c-47ca7a50ff06",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best Threshold = 0.146250\n"
     ]
    }
   ],
   "source": [
    "# calculate ROC curves\n",
    "fpr_stsf, tpr_stsf, thresholds_stsf = roc_curve(y_valid, probs_stsf)\n",
    "# get the best threshold using  Youden’s J statistic\n",
    "J_stsf = tpr_stsf - fpr_stsf\n",
    "ix_stsf = argmax(J_stsf)\n",
    "best_thre_stsf = thresholds_stsf[ix_stsf]\n",
    "print('Best Threshold = %f' % (best_thre_stsf))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "77bb1051-300a-4d9e-8be1-4639958fc7fa",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plot the roc curve for the model\n",
    "auc_stsf = round(roc_auc_score(y_valid, probs_stsf), 3)\n",
    "pyplot.plot([0,1], [0,1], linestyle='--', color = 'dimgrey', label='Baseline')\n",
    "pyplot.plot(fpr_stsf, tpr_stsf, marker='.', color = 'darkorange', markersize=5, label=\"STSF, AUC=\"+str(auc_stsf))\n",
    "pyplot.scatter(fpr_stsf[ix_stsf], tpr_stsf[ix_stsf], marker='*', s=70, color='dodgerblue', label='Optimal', zorder=2)\n",
    "# axis labels\n",
    "pyplot.xlabel('False Positive Rate')\n",
    "pyplot.ylabel('True Positive Rate')\n",
    "pyplot.legend()\n",
    "# displaying the title\n",
    "plt.title(\"STSF ROC curve\")\n",
    "# show the plot\n",
    "pyplot.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "e08edc7d-84bf-43ee-9a0b-1d79e832b76b",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Prediction run time of STSF -- 133.000000\n"
     ]
    }
   ],
   "source": [
    "# Get the model run time\n",
    "stsf_run_time = round((stsf_run_end - stsf_run_start))\n",
    "# evaluate scores for each threshold\n",
    "f_score_stsf = [f1_score(y_valid, to_labels(probs_stsf, t), average='weighted') for t in thresholds_cust]\n",
    "tss_score_stsf = [tss(y_valid, to_labels(probs_stsf, t)) for t in thresholds_cust]\n",
    "hss_score_stsf = [hss(y_valid, to_labels(probs_stsf, t)) for t in thresholds_cust]\n",
    "gss_score_stsf = [gss(y_valid, to_labels(probs_stsf, t)) for t in thresholds_cust]\n",
    "mcc_score_stsf = [matthews_corrcoef(y_valid, to_labels(probs_stsf, t)) for t in thresholds_cust]\n",
    "# get best threshold\n",
    "f1_ix_stsf = argmax(f_score_stsf)\n",
    "tss_ix_stsf = argmax(tss_score_stsf)\n",
    "hss_ix_stsf = argmax(hss_score_stsf)\n",
    "gss_ix_stsf = argmax(gss_score_stsf)\n",
    "mcc_ix_stsf = argmax(mcc_score_stsf)\n",
    "print('Prediction run time of STSF -- %f' % (stsf_run_time))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5125d620-569e-40a6-80dc-0c4ed971d654",
   "metadata": {},
   "source": [
    "### Visualizing the variation in scores for different thresholds."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "6970a891-16fd-47a5-a81b-92b4dc2ef78a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABEEAAAHgCAYAAABKGXM4AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguNCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8fJSN1AAAACXBIWXMAAA9hAAAPYQGoP6dpAAEAAElEQVR4nOzdd3hTZfvA8e9J0qZ7QemAUva0yCpLpggqsoqCgiiK8nODr+NFcLxOUBR4xfkKiiIqIshQURAEGSKUKXtTumlp6c5ocn5/lIamSQelE+7PdfWiPec5J/cJSZvceZ77VlRVVRFCCCGEEEIIIYS4xmlqOgAhhBBCCCGEEEKI6iBJECGEEEIIIYQQQlwXJAkihBBCCCGEEEKI64IkQYQQQgghhBBCCHFdkCSIEEIIIYQQQgghrguSBBFCCCGEEEIIIcR1QZIgQgghhBBCCCGEuC5IEkQIIYQQQgghhBDXBUmCCCGEEEIIIYQQ4rogSRAhhBDVbvv27dx77700adIEvV6Pt7c3zZs3Z9CgQbz22mscPHiwpkOsNb788ksUReGBBx6o6VAcNGnSBEVROHv2bJljFUW54q8vv/ySTZs2oSgK/fv3r/LrqSqHDh1i0qRJtGzZEnd3dzw9PWnSpAn9+vVj+vTp/P3339UWS//+/VEUhU2bNlXbbZbl7NmzZT4W/Pz8ajpMIYQQ1whdTQcghBDi+vLuu+8ydepUVFWlefPmDB48GG9vb2JjY/nrr79Yv349WVlZvPfeezUdqqhEEyZMcNh28uRJtm3bRlBQELfddpvD/hYtWpCfn18d4VWZ7777jgkTJmA2m2nYsCEDBgzA39+fpKQkdu/ezebNmzl+/DjLli2r6VBrBWePEwAPD49qjqTynT17lqZNmxIeHl6uxKEQQoiqIUkQIYQQ1Wb//v288MILaLVavv76a+655x67/QaDgZ9//hmj0VhDEdY+UVFR9OjRA19f35oO5ap8+eWXTrdt27aNNm3aON0PkJuby5EjR+rkm+CkpCQefvhhzGYzs2fPZsqUKWi1Wtv+/Px81q9fz6lTp6otpkWLFpGbm0vjxo2r7TavREmPAyGEEKKySBJECCFEtfnhhx+wWq2MGTPGIQEC4Obmxl133VUDkdVevr6+dT4BcjU8PDxo06ZNTYdRIT///DO5ubn06NGDZ555xmG/TqdzOgOmKtXW5IcQQghRXaQmiBBCiGpz/vx5AIKDg6/ouLLqYpS0v+j2lJQUHnnkERo2bIibmxutW7dmxowZmEymEm/3p59+YsiQIQQGBuLq6kpYWBgTJ07k9OnTDmOL1q7Izs5m6tSptGjRAr1ez8iRI/nggw9QFIX777+/xNtbtmwZiqJwyy23lOvaf/zxR2655RYaNWqEXq8nODiYrl278txzz5GSkuIw/vz580ydOpV27drh4eGBt7c3PXr0YMGCBaiq6jSm5ORkHnnkEUJCQnBzc6NNmzbMmDGj2paplFQTpOj2nJwc/v3vf9O0aVPc3Nxo1aoV8+bNs409cOAAd955J4GBgXh4eNCnT59S63BU5H4q6Txw5Y/3isZRtN7H+vXrGTx4MAEBASiKwr59+xzGFGe1Wlm0aBE333wzAQEB6PV6mjVrxpQpU0hOTnYa45U+BiuL1Wrlyy+/pE+fPvj5+dme088//zypqakO48t6fhY9b1XcB6+++ipNmzYFICYmxq7eSZMmTSr9/hFCCFEySYIIIYSoNmFhYUDBm/3ExMRqu920tDS6d+/Ojz/+SK9evRg0aBBxcXG8+OKLjBgxAovF4nDMI488wvDhw9mwYQOtW7dmxIgR+Pj4sHDhQjp37szOnTud3lZeXh79+vXj008/pV27dgwfPpzg4GDGjRuHi4sLP/74I9nZ2U6PXbRoEVByXYSiXnzxRe688062bNlC69atufPOO+nYsSMZGRnMnj3bYYnF3r176dChA7NmzSI3N5fBgwfTs2dPDh8+zKRJkxg/frzDbcTHx9O9e3c+++wzFEVhxIgRNGnShNdee43Ro0eXGWN1MJlMDBo0iC+++IIuXbrQp08fzp49y5QpU3jrrbfYvn07PXv25NChQ/Tr14/mzZuzdetWBg4cyNGjRx3OV5H7qSSFj/cNGzZw+PDhK7quq4ljyZIlDB48mJSUFG677TZ69+6NRlP6Sz6TycSIESOYMGEC0dHR3HjjjQwdOhRFUZg3bx5dunRxSP5d6WOwsqiqytixY3nwwQeJjo6mZ8+ejBgxgpycHN577z06d+7MyZMnnR5b0vOzqu+Djh07cueddwLg6enJhAkTbF8y+00IIaqZKoQQQlSTM2fOqJ6eniqgenh4qKNHj1bff/99devWrWpeXl6Jxy1cuFAF1AkTJlzR/sLtgNq3b181IyPDti8mJkZt1qyZCqjvv/++3XHz5s1TAbVjx47qiRMn7PZ98sknKqA2b95cNZvNtu0bN2603VaXLl3UlJQUhzhHjhypAuqXX37psC8lJUV1cXFRvby81Ozs7FKvLS8vT3Vzc1O9vb3VU6dOOZxr3759anJysu3n7OxstXHjxiqgzpkzR7VYLLZ9cXFxaufOnVVA/eKLL+zOM2LECBVQhw0bpubm5tq2Hzp0SG3QoIHtes+cOeMQQ3kUXlu/fv1KHFN4vxYfU/T+7tevn5qZmWnbt27dOhVQvby81PDwcHXWrFm2fVarVR0/frwKqA888IDdOSt6P5UkMzNTDQ4OVgHVxcVFHTp0qPruu++qGzZssIu3uIrG0a9fP9t9snDhQqfnLhyzceNGu+3PPPOMCqi33HKLmpiYaNtusVjU6dOn255Dha70MViaM2fO2OIujw8++EAF1LCwMLvnp8FgUMeNG6cCardu3eyOKc/zs6rvg8LrDA8PL9d1CiGEqBqSBBFCCFGtNm/erLZo0cL2hqTwy9XVVR02bJj6119/ORxztUkQRVHUgwcPOhy3ZMkSFVCbNWtm25afn682aNBA1Wg0DgmQQkOHDlUBdfXq1bZtRd9kbd++3elxK1asUAH15ptvdtj3/vvvl3oNRbefP39eBdQOHTo4vZ3iCpM6999/v9P9u3btUgG1c+fOtm1nz55VFUVRXV1d1djY2BLPWdNJEI1Gox49etThuI4dO6qA2rt3b4d9+/btUwG1SZMmdtsrcj+V5eDBg7akRdEvrVarDhgwQP3ll18cjqloHIUJjltvvbXEeJwlQVJSUlS9Xq/6+/urqampDsdYLBa1Q4cOKqD+888/qqpe+WOwNEWTICV9FY23MHm5ePFih3Olp6ervr6+KqBu2bLFtr2s52d13AeSBBFCiNpBlsMIIYSoVn369OHo0aP89ttvPP300/Ts2RN3d3dMJhM//fQTvXv35n//+1+l3uaNN95I+/btHbaPHj0avV7P6dOniY+PB2DPnj2cP3+eTp060aJFC6fn69evH4DTuhLBwcH06NHD6XF33HEH9evXZ9OmTcTGxtrtu5KlMIGBgYSHh/PPP//wzDPPcOTIkVLH//bbbwAlLmHp0qULXl5e7Nu3D4PBAMDmzZtRVZW+ffvSqFEjh2Puu+++MuOsDuHh4bRu3dphe+H/3aBBg0rcl5CQYLe9IvdTWdq3b8+uXbvYvHkzU6dOpV+/fvj4+GCxWNi4cSN33HEHL774YqXGERUVVa7YCm3cuBGj0cjNN99MvXr1HPZrNBr69OkDXH7MX+ljsLyKLhMp+lW4ZCUuLo7Tp0/j6urqtLiyn5+f7fr//PNPh/0lPT9r030ghBCiakkSRAghRLXTarXceuutzJ07l7/++ov09HR++ukn2rVrh9VqZfLkyQ5JgqtRUuFBjUZjq9sQFxcHwNmzZwHYvXu3XfHCol/PP/88cLnwZVGldd9wcXFh7NixWK1WFi9ebNt+5MgRdu/eTePGjR0KgJbk66+/JjQ0lLlz59KuXTsaNGjAqFGjWLBgAbm5uXZjC69p2LBhJV5TdnY2VquVCxcuANiSQiXdd35+frWia42zBA0U1F0oaX/hvuJFcStyP5WHoij06dOHt99+m02bNnHhwgU2btxIr169AJgxYwY7duyotDiutANM4e0tX768xNv76KOPAPvH/JU8Bsvryy+/dPpV2CGo8HHZuHFju3bDRTVv3txubFEl3Te16T4QQghRtaRFrhBCiBqn1+sZOnQoXbp0oUWLFuTm5vLbb78xadKkch1vtVpL3a8oSon71EtdNgrHFJ6rcePGDBgwoNTzdu/e3WGbu7t7qcdMmDCBDz74gK+//ppp06YB8NVXXwEFsytKi7WoPn36cOLECdauXcvatWvZsmULK1asYMWKFbzxxhts3ryZ8PBwu2saPnw4/v7+pZ5Xr9fb/VzeeGpKWQU/y9pf1NXcT1dCp9PRv39/1q1bR5s2bYiLi2P16tW2x9PVxlHWY7C4wttr164dkZGRpY4tOqPqSh6DlaXw+VrRMSXdN3XpPhBCCHF1JAkihBCi1ggJCaFNmzbs2bPHrr2mq6srAFlZWU6PO3fuXKnnLfyUtzir1WqbAdKwYUPgckePxo0b8+WXX15J+OXSpUsX2rdvz6FDh4iOjqZLly588803AKW2z3XGw8ODqKgo2/T/mJgYHnvsMX799VdeeOEFvvvuO6Dgmo4ePcrkyZMZOHBguc5deH+UdN9lZGSQkZFxRfHWdhW5n66Gp6cn3bp1Iy4uzu7xXt1xFD7mO3fufMWP+fI+BitL4cyec+fOYbFYnM4GOXPmDHD5MVwedek+EEIIcXVkOYwQQohqU9anuPn5+bY3MEWXMRS+mTl27JjT49atW1fqefft2+e0Reny5csxGo00a9bMdhuRkZHUq1ePnTt3VuqSnKIK634sWrSIP/74g7i4OHr27EmrVq2u6rzh4eG89NJLAPzzzz+27bfffjtQcL3l1bdvXxRF4c8//3S6rKDocp5rRUXup9KUZ9bCiRMnAPvHe2XHUZaBAwfi4uLC2rVrS2zfXF4lPQYrS6NGjWjatCkmk4klS5Y47M/IyGDFihXA5do95VEd90FhMjc/P/+qzi+EEOLqSBJECCFEtXnppZf417/+5bSAYFZWFpMmTSI9PR1vb2+GDBli2xcZGYm3tzeHDh2ye+NjtVp57bXX2L59e6m3q6oqTz75pN1Mkri4ONtylMmTJ9u2u7i48Morr2AymRgxYgT79u1zON/FixdZsGABycnJ5b72osaPH49Wq2XJkiV8/vnnwJXNAomJieHzzz93OjPmp59+AuxrH0yaNImwsDA+/fRT3n77bYxGo8NxO3bs4IcffrD9HB4ezvDhwzGZTDz55JN2BTiPHj3KG2+8Ue5464qK3E+l+eSTT3jooYfYtWuXwz6j0ciLL77IgQMH0Gq13HXXXVUWR1mCgoJ48sknSUlJISoqitOnTzuMSUpK4v3337e9gb/Sx2Bl+te//gXAtGnTOHXqlG174WP14sWLdOvWjd69e5f7nNVxHwQGBuLq6kpycjLp6enljk0IIUQlq8HONEIIIa4zU6ZMsbWpbNKkiTps2DB13Lhx6s0336z6+PiogKrX69Xly5c7HPv222/b2t326dNHjYqKUps2bap6eXmpTz31VKntZYcNG6Y2adJEDQwMVO+66y512LBhqqenpwqot912m5qfn+9we08//bTt9jp16qTeeeed6rBhw9Qbb7xRdXV1VQH1yJEjtvEltXItyW233Wa7L/R6vZqWluZ0nLMWuXv37rUd16NHD/Wee+5R77rrLrVVq1YqoHp5eak7duywO88///yjhoWFqYAaGBioDhw4UL377rvVvn37qqGhoSqg3n333XbHxMXFqeHh4SqghoaGqmPGjFFvv/12Va/XqyNGjFAbN25c4y1ySzp2woQJKqAuXLjQ6f7C+764itxPJZk7d67tdkJCQtTbb79dHTdunDp48GA1MDDQ1uJ33rx5lRKHs/a3xZU0xmQyqaNHj1YB1cXFRe3WrZvt/7t9+/aqVqtVATUvL09V1Yo9BktStEVueVgsFlusbm5u6u23367efffdaqNGjVRAbdSokUN76/I8P6vjPoiKirK1yR03bpz60EMPqVOnTi3XdQshhKgcUhNECCFEtXn55Zfp3r07v//+O3v37mXnzp1cuHABd3d3mjVrxoABA3jyySdt3R2Kmjp1Kn5+fnz44Yfs2LEDb29v+vfvz5tvvum0VW1RAQEB/P3330yfPp01a9aQlpZGeHg4EyZM4Pnnn3daV2Du3LmMGDGCTz75hG3btnHo0CE8PT0JCQlh7NixjBgxwmmc5TVhwgRbK9Rhw4aVWQCzqObNmzNnzhw2bdrEwYMHbbMJGjVqxNNPP82UKVMcurpERETwzz//8OGHH7Jq1Sp27tyJyWQiKCiIFi1a8OSTTzJmzBi7Yxo2bMjff//NK6+8wurVq1m1ahWNGzdm+vTpTJs2jZYtW1b4+muritxPJXnooYcIDw9n/fr17Ny5k/3795OSkoKrqyuNGzcmKiqKxx57jI4dO1ZpHOXh4uLC0qVLWblyJZ9//jnR0dHs3bsXX19fQkNDmTRpEiNHjsTNzQ2o2GOwsmg0GpYsWcLtt9/O559/ztatWzEajTRu3Jhnn32WqVOnEhgYeMXnrY77YP78+QQEBLB27VqWLl1Kfn4+4eHhvP3225Vx1wghhCgHRVXLsWBVCCGEqIO+/PJLHnzwQSZMmFAlRU6FEEIIIUTdIjVBhBBCCCGEEEIIcV2QJIgQQgghhBBCCCGuC5IEEUIIIYQQQgghxHVBaoIIIYQQQgghhBDiuiAzQYQQQgghhBBCCHFdkCSIEEIIIYQQQgghrgu6mg6gNrNarSQkJODt7Y2iKDUdjhBCCCGEEEIIIZxQVZWsrCxCQ0PRaEqe7yFJkFIkJCQQFhZW02EIIYQQQgghhBCiHGJjY2nUqFGJ+yUJUgpvb2+g4E708fGp4WiEEOLaZ8rJYXZoKADPJiTg6ulZwxEJIYQQQoi6IDMzk7CwMNv7+JJIEqQUhUtgfHx8JAkihBDVwKTV4nbpex8fH0mCCCGEEEKIK1JWKQspjCqEEEIIIYQQQojrgiRBhBBCCCGEEEIIcV2QJIgQQgghhBBCCCGuC5IEEUIIIYQQQgghxHVBkiBCCCGEEEIIIYS4Lkh3GCGEEEIIIYQQ1z2LxYLZbK7pMEQROp0OrVZbZseXKzpnpZ1JCCGEEEIIIYSoY1RVJSkpiYsXL9Z0KMIJrVZLgwYN8PX1rZRkiCRBhBBC1Bqunp78R1VrOgwhhBBCXEcKEyANGjTAw8OjUmcdiIpTVZX8/HwyMzNJTEwkLy+PkJCQqz6vJEGEEEIIIYQQQlyXLBaLLQFSr169mg5HOOHt7Y1eryc1NZUGDRqg1Wqv6nxSGFUIIYQQQgghxHWpsAaIh4dHDUciSuPp6YmqqpVSs0WSIEIIIYQQQgghrmuyBKZ2k8KoQohqkR4dTeKKFRiTktAHBxMSFYV/ZGRNhyWEEEIIIYQQFSIzQYQQTqVHR3Nm3jwMcXGoZjOGuDjOzJtHenR0TYcmhBBCCCGEEBUiSRAhhAOr0Uj8d98V/FDYqUNVQVFIWrmyxuISQgghhBBCiKshy2GEqEMqc3mK3bmCgvDv1g3VaiX7yBFyTp1Czc93PEhVMSQmXuVVCCGEEEIIIUTNkJkgQtQRlbk8Je3vvwvOFRtrO1fijz+StHIl2ceOFSRANE5+PSgKbpXQm1sIIYQQQghRvRRFKffX2bNnAcjKymLGjBl06tQJPz8/vLy8aNq0KSNHjmTBggUOt7Ft2zZGjBhBkyZN0Ov1BAUF0bVrV6ZMmcLp06er+Yqdk5kgQtQRiStWFHxTdHkKELd4MfoGDXBv1AhFq3U6W8S3UydyT58m68gRso8cIevQIae3ofXwoOE99+DVrh15MTGc+eADUBS72wyJiqrqSxVCCCGEEEJUsq+//tru5yNHjjBjxgyioqIYNWqU3b7AwECysrKIjIzk1KlTjB49mgcffBBXV1dOnz7N77//zvvvv8/DDz9sO+azzz7jkUceoUWLFjzwwAM0atSIlJQUDh48yMKFC+nTpw/NmjWrlmstjSRBhKgDMg8cwBAb63SfOS2Noy+9hOLqimu9ehiLLFcxxMZyZt48FJ3O+fKWYqxmM/UHDADALSgIJk8maeVK8uLiwGpFo9fj1bp15VyUEE7kG42sfeYZAG6dMwedXl/DEQkhhBBCXBvGjx9v9/OmTZuYMWMGHTp0cNgHMGfOHI4dO8a8efN46qmnHPbHxcXZvs/Pz+eFF14gPDycPXv24O3tbTc2Ly+PvLy8SrqSqyNJECFqsdyYGOKXLCHr4MESx2j0ehStFkturl0CpCg1Px+dtzdebdrg3bYt59etw5icfHmGBzhd6uIfGYl/ZCSqxcLRV14h79w5En74gcYTJ1bK9QlRnDU/n10ffwzAoFmzQJIgQgghhKjDKrOmX3U7fvw4AAMufUhaXKNGjWzfp6amkp6ezs033+yQAAFwd3fH3d29agK9QpIEEaKWKPoL0jUwEBcfH7KPHQNVRdFq8Y6IIHPfvsvLUy792+TRR/Ht3BljcjKHp00Di8Xh3IpOR8SHH6JcqvOh8/PjzLx5DucqaamLotUSdv/9HH/zTVI3baJe//541oKpbEIIIYQQQlQFVVWxGo1XdY6Le/YQ88knttfahTX9rI89hl/nzhU+r0avR1GUq4qtPAqXrixcuJB33nkHna7k9EGDBg3w8vJi8+bNHDt2jNa1ePa4JEGEqAUKi54W/oI0JiRgTEgAwL9HD0JHj0bfoAHp0dEkrVyJITERt5AQQqKi8OvaFQC3kBDcQkMxxMU5zvAIDbUlQKBghkfhUhdn53LGq3VrAm66ibRt24hdtIjWr7xid04hhBBCCCGuFVajkf2TJlXOyYrV9Iv55BNiruJ0N86fj9bN7erjKsOkSZP48MMPmTNnDosXL6ZPnz5069aNm266iZ49e6Ip8l5Ao9Hw6quv8txzz9G+fXs6d+5Mz5496d69OzfffDPBwcFVHm95SRJEiFog8ccfC74pmrwA9EFBNH3iCdvPhctTShISFVXuGR5lncuZhvfcw8Xdu8k9dYq0rVup17fvFR0vhBBCCCGEqBv8/f3ZvXs3s2fPZtmyZSxfvpzly5cD0LRpUz799FMGDx5sG//ss8/SqlUrPvroIzZv3kz0pS6WWq2WCRMm8MEHH+Dh4VEj11KUJEGEqEEWg4GU9esLZm84YUpLu6LzVWSGx5Vw8fMjZORI4pcsIf777/Ht0gWdp2elnBvq9ppJIYQQQghx7dDo9dw4f/5VnePYa69hiI93nKXdsCGt//Ofq4qtugQGBvL222/z9ttvk5qaSnR0NEuWLOHrr78mKiqK/fv306JFC9v4YcOGMWzYMPLz8zl27BgbNmxg7ty5fPHFF+h0Ov73v/9VW+wlkSSIEDXAajSSsn49yb/8Qn5WlvNBTgqVlkdFZnhcicBbbyV182aMCQkkrlhBmJNK0hVRfElQ4ZpJJk+WRIgQQgghhKhWiqJc9ZKTkFGjnM7SDr3zzmpZzlLZ6tevz+23387tt99Ow4YNmTlzJkuWLOGll15yGKvT6Wjfvj3t27dn7NixtGzZkq+++oqPP/4YrVZbA9EXia1Gb12I64RthkNiIlovL6xGI9ZLLaL0DRrgc+ONpPz+e7kLldYkjU5H2PjxnJw1i5Tff6d+v364h4WVekxpMzxUVcWQkEBcYd/yYmsm45cswatlS3S+viiKIrNFhBBCCCFEnVDVs7RrUs+ePQGIj48vc2xgYCDNmzdnz549pKamEhQUVNXhlUqSIEJUMdsMh0vyL14EQOfjQ8O77ybgpptQtFq82ratM78gfSIi8IuM5GJ0NLGLFtFy+vQSK1SXNMMja8AALHl5ZB05Qn5GRom3ZTp/ngNPPYXW0xOdj49dG2CZLSKEEEIIIWqzqp6lXZW2b99O27Zt8fPzc9i3atUqANq1awdAbm4uO3fupH///g5jT5w4weHDh6lfvz6BgYFVGXK5SBJEiCpmK3pajM7X166waF37Bdlw3Dgy9u8n++hR0v/+m4BL2eDiElesKPim2AyP1I0bbWMUFxcUjcZpGzJFp0O1WLDk5GDJybHfeWnWTNLKlXXqvhNCCCGEEKK2++abb1i4cCFDhgyhe/fu1KtXjwsXLrBmzRo2btxIu3btmDhxIlCQBBkwYADt2rVjyJAhtGzZElVVOXr0KIsWLcJgMPDRRx/ZdZSpKZIEEaIKWXJzC4ohOWFMSqrmaCqXvn59gocNI3H5cs5++ikxn32G/tIMFo8mTcg+coSsI0cwxMY6P4GiEDxyJN5t2+LZvDkZ+/c7XTPZ9Ikn8OnQAUNiIkf/8x+wWOzPo6oYiswOEXWboigEXvpEoaTZRUIIIYQQouo9+uij+Pn5sXHjRubMmUNqaip6vZ4WLVrwn//8h2eeeQbPS00S/Pz8+OKLL1i3bh2rV68mMTERg8FAYGAg/fr146mnnmLAgAE1fEUFFFUt1pNT2GRmZuLr60tGRgY+Pj41HY6oY8wZGZx8913yYpx0AVcU3MPCaPvWW9UfWCVK276dsx9/fOUHlnD96dHRpS4JOjx9ekEnnWK/tnS+vkR88IG8aRZCCCGEEFfEYDBw5swZmjZtilsdLFZ6vSjP/1N537/LTBAhqoAxJYWTs2ZhTEpC4+5eUAS1DhQ9vVJJP/1U4j7PFi3watsWRaslaeXKcl1/WUuCQqKi7GeLXJKfkUHsl18Sdv/9KDVcbVoIIYQQQghRe0kSRIhKlhcfz8l33sGcno5r/fq0mDqVvNjYOlP09EqUtKRH0ensep+7N25cKdfvrMK2R5MmXNiyhdQ//sCcnk7TJ56o1t7pQgghhBBCiLpDkiBCVKKcU6c4+d57WLKzcWvYkBb//jeuAQG4BQdfk4U79cHBjstTFAW30FC7cZVZ9NXZuXw6duTsJ5+QsXcvx2fOpPkzz+AiS9iEEEIIIYQQxUgSRIirkB4dTeKKFRiTknDx88Ocno6an49Hs2a0eO45dN7eNR1ilXJYnlJDS338IyNx8fXl1Ny55J46xZHp09F6eGBKTUUfHExIVNQ1mYQSQgghhBBCXJma708jRB2VHh3NmXnzMMTFoZrNmFJSUPPzcQsLo+W0add8AgQKkg9NJ0/GPSwMxcUF97Awmk2ZUiNLfbxataL1yy+j8/YmPyMDY2IiqtmMIS6OM/PmkR4dXe0xCSGEEEIIIWoXmQkiRAUlrljhUKCzkPY6qixdmUtdrpZbaGhBEiQr6/LGSzNUklaurDVxCiGEEEIIIWqGJEGEqCBjUpLTBEhJxUJF9TCmpDhuVFUMCQnVH4y4ImlZFrIMjs8pb3cNAV6aco8RQgghhBCiJJIEEaKC9MHBGGJj7TcqCm4hITUTkABKKNYKqFYrmQcP4nPDDTUUmSiN2aLy1rJMMvMcExw+Hgpv3+cHUOYYF61S1aEKIYQQQog6TD42E6KCHN5M11BRUGEvJCrqcpFWuPyv1crJd94h7rvvsObn11yAVykty0JMSr7DV1q29YrG1DY6DQR4aSiewlAAPw8NqCpaRS1xTICnBp38RRNCCCGEEGWQmSBCVEB+djZp27YBoPPxwZKXh1tICCFRUTVSFFRc5h8ZCZMnk7RyJYbERNxCQggaOpTso0dJ/eMPzq9ZQ9ahQzR94ok6N2unMmdL1LZlJYqi0Kt9LmfXWQn8+G4AUh7/HtXVnXOpFh7/7CKKAloFiketAiO6uaMol9Mjte36hBBCCCFE7SBJECEqIP6778jPzMQtNJQ2b76JxsWlpkMSRTgr1hrQsyc+ERHEfP45eTExHJk+vaCIanZ2lbfRraxaFzoNeLlbyMxTwG4+hIqXm9U2E6KsMeVJplR3ouRkoplPN5zCVw1Hf6IgwYjVYjdGVSHfMRwAFm/OpHdbT7q1dMXfSyPLZoQQQgghhFOSBBHiCmUePMiFzZtBUWj80EOSAKlD/Lp2xaN5c06++y6G2FjM6ekAGGJjOTNvHkyeXOmJkMqavaHVqOyKOcX2pF8Id72/2CiF06nnuf/jS8VfVRfcNEEOY/af38y/f/DC202D0dIS8ORqEyXlVVJCxWJR2XDAyM4TJny07bCYL1yOWFHYa5jMReseNLgWfCmu+Gsiaat/+fI5VBMXslxZtTOPVTvzaBKoQastuLKityjLZoQQQgghhCRBhLgCVqORcwsXAhA4cCBerVrVcETiSrn6+5e4L/HHHyuUBEmPjiZxxQqMSUkOs0p0GnAni0zVA5Qi775VK+5qLjqNH1DymFxDJk8s2oMxNwRXJdhJAqSAXdKjhNxEoK4fF1PhYolXonA45QCvrvQkLMATjbYeBX8mSp51Uh6lJVQun9VKYv5qzho+pPGlbZmWIyjux7m10c2kGdJIy00jITOBRPNqGlpG4aNtT6blEHuNT9CzwcO08b6HM0k6zqY4r32iAiO62y+bEUIIIYQQ1xdJgghxBRJ+/BHT+fO4BAQQOmZMTYcjKqikNsaGuDhS1q+nfv/+KLry/XpMj44umEVyqTCuIS6OM/Pmkf/gg3g0aUJ+ZiZr09+lo9sH9gcqGg5m7eKJj3cCkGDREKjr7zAm3+oHeZ1wVUBV88k3nyCbBPxdB9qGpeR8wMi4HBSNFgDVamFlI08CPZ+yjblgWEFTqw41tBl5JoWcPBWdtTka9HZJAT9tBxISISGxpCtWWJMwhdeUxbYtZS2bKSx6mpVncajnAXDRsofjpjm4uqVQ39vNtt3qtY5Qz1DmDJ2Dq9a14NpUlajFUZxK+ZhWyrOcMn+MhRy2nn+fv1I+4Nbmd9Ij8P+ITazH6fNFl9OoNAnU0T5MZm4JIYQQ4vq1adMmBgwYwMyZM3nhhRecjvHy8qJr165s2rTJtm3NmjX897//5fDhw6SkpBAQEEDTpk3p3bs3//73v6lfv75t7Pnz53nvvfdYs2YNMTExaDQaGjRoQJcuXRgzZgyjRo2q6ssslSRBhCin3LNnOf/rrwA0fuABtO7uNRyRqKiS2ugCxH71FefXriV0zBj8unYtddaA1WQifsmSgh8Kz3Xp39hLM4YMGi0dW92AVW9Go9i/AQ/U9cds+97x/KpqxZB/iHrZx2mZdoaOFxJxsZp4ucUBPHwbondpg9F8lNy85bRNj8BFLZieYVasfFvvAEbXQbYx2dnvM+ZMRyInj0Xj6oqqqoz58iX88p613V6q5hvaBTXlfKaF7Fw9WOrhqTRFUeynfXRwf425v57kjhvDCG/gUuaymew8lcZBeZxNcXUYc9b0BYl8yUPdH2JSt0lgMDN7cgAAX49/FcXd1ZYAATBbC+6xRoF5DO1wnKX/5OFvbkZDn4ZsObuFX0/+wK8nf6BzaGe6NX+M46c6XjpSoVHwBRIys2no29DxzhZCCCGEEE5Nnz6dmTNn0rFjRx5//HGCgoJISEhg7969zJs3jzFjxtiSILGxsURGRpKVlcW9997LY489BsDJkyf55ZdfyM7OliSIEHWBmp9PzIIFoKr49+iBb6dONR2SuAohUVF2szcK/63Xvz8Ze/ZgTErizLx5eLZogU/HjqTv2FGw1CUoCP8ePcBqJevIEXJOniRL8STPs5HjjViM/NmsLYneA9Fp/Bz6kaeZ1tLIkIOLvx8A5vSLxLl5EuB6q23M+cyXeDNZT9i949Hou6PV64mZP583TkKSz3q2NfHjprPrCc7qgHdIQ1pMnQrAiXfecRyTGYGLqnLwX/+iwa234tO/LwbjToz5x9DrWmPMP0a+62r+M/wXW9Ih15TL2K9epYH58qcEFtWAFn8On4HDZ3JQtFnote44+3OSn68ydVEaWXkKYJ8AUVUrWdZjnM7/hL8e/Ysgr4LlPCZzjm2Moih2CRAAV60rS8ctxUXjgqIojO0wFrPVjKvWlaMpR5kfPZ+fjvzEnoQ97EmYRFf9Iny0bQHY9I8L70c/yZ5nVpT42BBCCCGEEJelpKQwa9YsunfvztatW9EVmy2dkZGBVqu1/fzuu++SnJzM6tWrGTZsmN3YuXPnEhcXVy1xl0aSIEKUQ/Jvv5EXE4PWy4tG48fXdDjiKjlro1vY3tgybhzJa9Zwfs0ack6eJOfkSdtxhrg4Epcts/1sUbQs6/wsea4+DrehqlYURYMOMKoJnM/7mgYuI2wzM7IyZ/LSbZ8R2K0HAOd3bGf82kfw9A23jTGY/yJk3Gd2dUpC776bM/PmEZZ5knv2z7yUwFEIHT0a14CCGRShd93lZIwGnZcX+ZmZJPzwA4krV/JSfgOSfH5nWxPfS4mSBuR02Y9rYT0TrQ5Ffxyz5TQu1maYNaeJ175Me/9RnEv2xZce6CzeGCwOlw9ArglAQVUtZKsnMVjPE6jrA4CiaDht/pg5Q+bYEiDlVTQxUjRR0iawDbOHzOaZm57hxd9fZMvZLZwyf0gr5TksqgkfbWu6ef6Pk4lmWoTIshghhBBCVI9tMdt4fcPrvDLwFW4Kv6mmw7kip06dwmKx0Lt3b4cECICvr6/dz8ePHwdgwIABTs/XqJGTDw+rmSRBhCiDISmJxB9/BKDRuHG4FHuii7rJWRtdAK27O6F33kn9m2/myLRpWHJyHMZo3N1pOGYMXm3b4bMun7xcq31BUwre5BvUGNo3T+HJAb3J3TeS3Wv3s9HTnwE5/9C1SAIEoEH3nizmf6WOKYy7pAROWWN8O3YkbccOkn/5BUNsLC5oCMs4UZAoAVC0JK1cabtfXLWuLL13KScS4PutudzduwMtQ1fhqnXFbDGz5cx2ftp/jBPxbgRrb0NRLtUkUVXyySLG/CVZ6iE83NMJ9a1P7IUT6C3/tRU0bVg/h+Fth1fsP7AUDX0bsvDOhQz/ejiHz+9kh2EMGvT09Pofekt73v85i6eHedM8WBIhQgghhHBOVVXyzHlXdY6EzATSDem8/efbnEw7ydt/vs0rN7+Cv5s/oT6hFT6vu8vVF3rPzc0lNTW1zHHNmjUD4Oeff+aZZ54hNLT0uAvHz58/n6effrpWFqRXVNXJovga9t133/Hee+9x+PBhPD09GTRoEG+//Tbh4eFlHquqKosXL+bjjz/m+PHjmM1mwsPDGTt2LJMnT8bLy6vccWRmZuLr60tGRgY+Po6f9IprW2HHj8LaEW5hYbR9661a+UQWVWPvxImoZrPDdsXFhU5ffAFA5Jx7HYueAmdMC1j52BN46T2rPM4rpaoq+x58ENXiOIWj6LWV1/KDy3ln3WpudJtr27bP8BSP9r2Z+zvdj4vWBZPFxJhvx+CS345QHiWBTzHrDrN03FK7mR2mnBxmXvo9PS07G1fPit1/hbdnUS2czz5Pam4qHjpf7m28nlNJ4OaCJEKEEEIIgcFg4MyZMzRt2hQ3t8sF2nNNuUTMi6jByEp2YPIBPFw9KnRsYWHUsvTr189WGPWpp57iww8/xNXVle7du9OjRw+6d+/OwIED8fPzszvu9OnTdOrUiczMTMLCwujTpw+RkZH06dOHLl26VChmKPn/qajyvn+vdTNBPvzwQ5566iluuukm5s6dS2pqKv/973/ZvHkz0dHRZWaepk+fzttvv83NN9/Ma6+9hlar5ffff+fFF19k7dq1/Pnnn9V0JaIus3X8KMIQG8vFXbsq1EJV1E1OC6gqCm4hIbYfH+t1N9t256NRCn6dqqqFLOtRnhzUulYmQKBgCYk+NNRpcVh90JUtTQEY1X4UX+/5msyMQ3azPCZ2mWhLGhav5aGqb9lqeRSl0Wppd9ddtu8rqujtxWXEMeqbUaTlpXHU8gLtQ9/leIKFuauzGNvHk0b17W+nsKuNEEIIIcS16qGHHuKee+5xuq94LY958+YRGRnJ559/zt9//82WLVsA0Ov1TJkyhRkzZtjqgjRr1oz9+/fz7rvvsmrVKr799lu+/fZbACIiIli4cOFVJUMqQ62aCXLhwgWaNGlCq1at2LFjh23N0a5du+jWrRsTJ05kwYIFJR6fn5+Pr68vbdq0ITo6Go3m8ovYkSNHsmrVKo4cOUKbNm3KFY/MBLl+HZo6FWNCgv1GRcH90mwQcX0o3v628N9mU6bg17UradlG/v3NWRRLoN1xF91ns/SBN2v1rCGHa7tEHxxM61dfRXcFMzDKO8ujJv2T9A/jlowjLz+PO9uNpZ7xeU4mOi9mUtjVxkVbe///hBBCCFE5SpphUBnLYVRVZez3Yzl0/pBtW/sG7fnu7u+u6nXi1SyHqWiL3EJGo5FDhw6xdu1a5s6dS0pKCjNmzGDatGlOz5WYmMj27dv56quvWL16NcHBwRw6dIiAS7XsyqsyZ4LUqo+6Vq1aRXZ2NpMnT7YrutK1a1f69u3L0qVLMZlMJR5vNpvJy8sjODjYLgEC2GaQeHhUbNqQuD7k5+QQv2SJYwIEQFUxJCZWf1CixvhHRtJ08mTcw8JQXFxwDwuzJUAMJisvfn8CxRJInjUBk3IWALPmNGbtIVsr19qq+LXpg4LQ6PUYk5I4+c475GdllftchbMulj7wFu8/2JilD7xVqxIgAB2CO/DBsA/QKBqWH/4O13qL0TtZCaMAAZ4adLXqr6MQQgghqpuiKHi4elzVl4vOBY2iISIogjcGvUFEUAQaRYOLzuWqzluTH7Tp9Xo6d+7MtGnT2LJlC4qi8Pnnn5c4PiQkhFGjRrFq1SrGjh1LUlISa9asqcaIHdWq5TA7d+4EoFevXg77evXqxZ9//snRo0fp0KGD0+Pd3d3p1asXv/32G7NmzWLUqFHodDp+//13Fi5cyMMPP0zjxo1LvH2j0YjRaLT9nJmZeZVXJOoKq9lMyvr1JK1ejSU72/mgYssgRN2VlmUhy+A4Cc7ZMghnBVQtVpXpSw+Sb2yEWc1gZO802tbrUaR4aO1KAJSk+LXlxsRw8p13yD1zhhMzZ9Ji6tRyFwIuqWNLbTKg+QBev+V1Xvr9JT7cOYdnurRj35Eb7caowIjuV19sTAghhBCi+HLgsR3GOl0OXFe1bt0af39/4uPjyzW+Z8+efPfdd+UeX1VqVRKk8M5w1jancFtcXFyJSRCAb7/9lgkTJjB16lSmTp0KgEaj4dVXX+Xll18u9fZnzpzJa6+9VtHwRR1SWPTUmJSEzscH1Wwm/1LSy61hQ3y7dCF59WqHZRAhUVE1HLm4WmaLylvLMsnMc0yClGcZhKqqvLnqIFmZjbCoRrpFHOGujkMAeH1s3f6D5hEeTssXX+TEzJnkxcZyYsYMWk6bhkuxgld12dgbxxKfGc8nOz5h3t5HGdlgA2mZl2cIhvpraB8mxVKFEEIIUTnqwgdFpUlKSiIxMZFOnTo57NuyZQtpaWl07tzZtm3jxo306NEDd3d3u7FWq5WffvoJgHbt2lVt0GWoVUmQ3NxcoGCKTXGF634Kx5TEw8OD1q1b07hxY2677TY0Gg0rV67klVdewWKx8Oqrr5Z47LRp03jmmWdsPxdWtBXXluJFT80XLgCg9fCg4dix1OvTB0WrxaNJk1LbkIq6SaeBAC8NWXkWiqZByrsM4pMNR4lLbIiqWmnUeAuP9x1TleFWO/eGDWn10kucmDkTQ0ICR156Ca2nJ6aUFPTBwYRERdX54sDP9n6WxKxEVh5eyV9p/6GN7l3bvvQcK0kXrYT4V7woqxBCCCHEtSIuLo7IyEgiIyMZNGgQTZs2xWQysX//fr755htcXFyYMWOGbfzs2bPZtm0bQ4cOpUuXLvj6+pKUlMTy5cvZvXs3AwYM4I477qjBK6plSZDCeh1Go9Ehc5SXl2c3xpnc3Fx69epFly5d+O6772zb7777btzd3Xn99dcZOXIkHTt2dHq8Xq93moAR15bEFSucbncJCKB+//62n50tgxB1n6Io9L4hl7Mb7Z/rKtD7hlwU5fLyj+LLZjYcimfv8YLuKfqAdbw61HlF7brOLTiYVi++yLFXXyU/I4P8jAwADHFxBQnEyZPr9HNDURRm3jqTcxfPsSdhE6Gaw/ho2mHFSJ5Jz7srL/LCKD8a+EoiRAghhBDXtzZt2vDRRx/x+++/s2TJEpKTkzGbzYSEhDBy5EieffZZu1kiL730Ej/88AObN29m3bp1pKWl4enpSdu2bZk9ezZPPPGEQ/3O6larkiANGzYECrJNLVu2tNtX2lKZQsuWLePEiRO8/fbbDvvuvvtuFi5cyJ9//lliEkRcH4wlFDc1JidXcySipjz8y0B6uf+Mq+Jnt/2j9ceITzdzW4dGeLsrTpbNeAOgYuSdUaPQKNdu9Ux9gwZovb3tC6ReWhqWtHJlnU6CQMHU1D0JewA4ZfqIVq7Pctr0KU1dH4W8ZsxZncXzI72p5y2JECGEEEJcO/r3709ZDWKzi9RI9PLy4vHHH+fxxx8v1/l79OhBjx49rirGqlarXsFHXnpR/ddffzns++uvv/Dy8iq1vW1hosRsduzKULgtPz+/MkIVdZjGWUslKXp63UhMtzDYf7VDAkRVrfhqO7BxnxdTF13k2a+PoZJP8eogKiph9fR4ul77s8ZMKSmOG1UVg7PuSXXQnCFzAEi37mSH4W5SrBvZZ3gcT/dcLmRZmbM6i4s51hqOUgghhBBCVKZalQQZMWIEHh4ezJs3zy5ZsWvXLjZv3syYMWNwdS0oJJOYmMjRo0ftaoQUFlj56quvHM79xRdfANCtW7eqvARRy2UfO+bY/UWKntYZaVkWYlLyHb7Sssv3RnXnCSOvL03DZAzAqF4gx3oWgEzLIc7oJoH3SjKs+1BVK3m5DcjK01I8T66g0L+D8broHqIPDi54fhSj5ucTM38+prS0Goiq8gxvO5yIoAi7bYE+brx8Vwj1vDWcz7Dy7soMDseaKvyYE0IIIYQQtYuiljUXppq9//77PP3009x0003cd999pKamMnfuXFxcXNi1a5dtycwDDzzAV199xcaNG+l/qY6DxWKhV69e7Ny5kz59+jBq1CgURWHlypVs2rSJoUOH2irSlkdmZia+vr5kZGTg4+NTFZcrqpHVZOLIiy9iTErCq107LNnZUvS0DjFbVF5YdLHMri7O2t/mW1Q2HTLy9zETAOmWXcTwNuFe3WnIYyTwKWbdYZaOW4rBbODnw5v549B5Mi6G46e9vMZRVS1kWY+yy/gAp547VbUXXAvYiggX65JUSHF1pcFttxF8xx1kHjpk67hUVwqomiwmxnxbUNg2xDuEdSfXoaCwdOxSwrxvZNaPmVzMdf4nsjydhIQQQghR+xkMBs6cOUPTpk1tzThE7VOe/6fyvn+vVTVBAKZMmUL9+vWZPXs2Tz/9NB4eHgwaNIiZM2faEiAl0Wq1bNy4kQ8//JDvvvuOV199FYPBQIsWLXjrrbd47rnnqukqRG2UsHw5xqQkXPz9aTZ5MjpPz5oOSVyB8nR1Ka39baGz5i/o0jqNrwf+gpvODUVRUNW3bD3bXbWujOs8lHGd4bv93zHvj5e4Qf9mwW0pWk6bP7Uto7jW+UdGwuTJDl2SdL6+xC9ZQs7x4ySvXk3KunVYDQZbkuRqCqiacnJ4v0kTAKacPYtrFT5PXbWuLB23FBeNCyoqj654lA2nN/DMmmdYdd8qnhnhzatLMrEWeziVt5OQEEIIIYSofWrdTJDaRGaCXDtyTp7k2Ouvg6rS/Jln8HXS51rUfgfPmXj/52yH7SO6udEh3BVfD4UP1mRzLsXisIzFquZzwPgsd3WN4Pk+z5drOYuqqkR9HYU+41l8tO3JtBzC6DubFfetuC6Ww5RGVVUydu8m/vvvMSYlOQ5QFNzDwmj71ltXdF5TTg4zvbwAmJadXaVJkOIyDBmM+HoEsRmxDGw+kE9HfsqWQyYWb3ZszT5lqBc3NHatttiEEEIIUTVkJkjdcE3PBBGislnNZmLmzwdVJeCmmyQBUoe1bajDQ6+Qa7RPcazaaWDVTgNQ8Cm9s8zuEdPrPNS7N490e6Tct2e2mkEBq9eveBDORX4FBduskeuZoij4de2Kb8eO7H3oIbAWq5FRBwuo+rr58uHwDxn97Wg2nNrA/Oj5/F/k/7HhgIHE9MvXpwB/HzOiqtC2kQuZuVaHJVgA3u4aArxkuogQQgghRG0iSRBxzUtcuRJDQgI6X18ajR9f0+GIClJVlf+tTyXXaP9rK8jPhKLoyDVoycpTHRIgqmoly3qU+3recEUJELBfLlF82YwooOh0uDVsiCEuzq5eCBQUUD01Zw6Bgwbh3b49F3fvLrNuyMXdu23fH/3Pfwi7++5qrS1yQ9ANvHLzK7z0+0vM3jKbjiEdGXNTJ7sZSCqw44SZHSfMeOrBlA9mi+O5pG6IEEIIIUTtI0kQcU3LPXOG5J9/BiBswgR0l6bZi7pFVVWW/pXL3lM6VNVCnpqAhyaMTMsh/kh4ACj4FB805Bl1BGoH0Nr1eQAURcNp8yfs2vo3j/V49Ipvu2jCQ1EUSYA4ERIVZV9AtYiMvXvJ2LsXnZ8f+RcvOtQNsU6ahEfz5uRnZZGxZw8JRYpXG+LjK1xb5Grc0+EedsXvYuXhlUz5eQqr7/uJJoGunE2xEB6o5e6b3Ik+aWbXKRNZJdSfkbohQgghhBC1kyRBxDXLmp/P2fnzwWrFr3v3Wt+pQpTsp10G1u83AnDE9AZGNYVWrs9yyvyxbUyGIcP2fXz+UkK0Q2x1PNKsf183xUxrQkkFVN0aNiRl/XoubNlSkACBy0mSS//GzJ9f5vkTly2r1uevoii8ccsbHEo+xIkLJ3j6lylM6/kFP2wzMKqHBy1DXWgZ6srdvT04GpfP2n15HInLtzuHCgT7azmfYSXITwvgtHMR2C+bKc8YIYQQQghRcZIEEdes5NWrMcTGovP2Juz++2s6HFFB6/cb+Ck6DwCfwE0kxfwCwA7D3QC0qt+KD4d9aCtUqqoqT/70JKfSPqaVUpAoiQiKYHjb4TVzAdcJ/8hIp4mKsPvuI3T0aPY/8ohj3ZBLtB4e6Hx8MCYnO91vSEjg7GefEXTHHbiX0SWssni4evDxiI8Z+fVIdsTu4HPPFziSd4Rbra8ANxXErVFo39iFdmE63lqW6VCQ9+/jJv4+biI8UEvXFq6s3Wsg20mCo3DZDFBidyNZWiOEEEIIUTkkCSKuOenR0SQsXWrrWOF/0024SHefOmnrESPfbyvozNGyyRn+d7hgiUsTvyY8FPkQS/9ZCkCYX5htmYrJYkKv1dMoMI+hHY6z9J+CBIrU8qg5Wjc353VDFAW3Ro1oN2MGAIenTyf7zBmn50jbsoW0LVvw7dQJj+bNSd+xo9TaIpWhWUAznuvzHK/98Ro/HS1YpjNr8yzeHPQmAe4BNPRteOkyFEZ2d7erGzL4Rj3xaVaOxJmJSbEQk5JX4u24uygcjTMVfO+qOCRBZGmNEEIIIUTlkRa5pZAWuXVPenR0QQ2BYppWc00BceWKLwM4HGvmx78L3jh2bpnNB4duw5hv5InuT/Cv3v+6VKhUdZrcMFlMRYqZOh8jqpftuVlYN+TSv82mTMGva1fbmBNz5vDrkiUA3H7PPeh0OkLHjCH39OmCoqnF/2RdOk9VPcebv9e8xH2nnjtl+15VVWYsy+RsioUmgVqm3+WDoihk5lrZfcrEjhMmTiXll3iuskhLXiGEEKJqSIvcukFa5ApRgsQVKxw3KgpJK1dKEqQWM1vUEpcBuGhhyakHMOYbGdBsAE/3ftq29KWkQqVSzLT2KaluSGECpHBMk8ceg0tJEPdGjQi7+27bGENiIsdeew1LTs7lE19KqFTVc3zOkDk8s+YZh+1BXkEsiF7AyHYjqe9ZH0VRiOrhwZKtOUT18LA9Rn08NAyIcGNAhBvnM/J5d2UWF3MuP85dtBDkp7FbzpV80WrXbSY8UEv7MJdKvzYhhBBCiOuRJEHENcWQkOC4UVUxJCZWfzCi3HQaCPDSkJVncWhxa+QccdkxNPZrzJwhc9AosiagriqpbkhR9Xr0oN9//gNAu+nT0bpeTmC5hYRgNZkcD6rC5/jwtsNZuHshB5IP2LYpKCRnJzPzz5nM2jyLAc0GcOcNd6LX6tmRN8OubkhRDXx1TBjgabds5vHbHWd4HDxnshvTuqHOliQRQgghhKgMBoOBL774gmXLlnHgwAEuXryIp6cnLVu2ZMCAAUycOJE2bdrYHXP+/Hnee+891qxZQ0xMDBqNhgYNGtClSxfGjBnDqFGj7MavWbOG//73vxw+fJiUlBQCAgJo2rQpvXv35t///jf169evzku2kSSIuGZkHToEFovjDkXBLSSk+gMS5aYoCiOK1VQo9E/Ou7jr3Pl0xKf4uMmytGud1tWV/q++WuJ+fXCwY20RCuqOqKpa6ckCs9UMQERQBGM6jGHpP0uxqlZGR4xm1eFV7E3cy/pT61l/aj1aRYtFtfDGH2/w7u3v2tUNKdQ+zIUmgVrbshlnMzyKjgHYfMjIzRFu1PPWVuq1CSGEEOL6dPr0aYYOHcqRI0fo168f//rXvwgJCSE7O5t9+/axcOFCZs+ezblz52h4qSh9bGwskZGRZGVlce+99/LYY48BcPLkSX755Reys7PtkiDTp09n5syZdOzYkccff5ygoCASEhLYu3cv8+bNY8yYMZIEEeJqGBITOV20FkixugMhUVE1F5wok6qqJKUXT2CpZFoOk2b9m/eHvE/rwNY1EpuoXUKiouxri1ySn5VFwtKlhI4ZU6mJEFetK0vHLbXVmBnbYaytxsx9ne7j5IWT3LrwVgAsasFj+MSFE4xcPBKA488cR6u5nLwoadlMUYVjvtuSgwIkXrTyxYYcnh3hjUZmhAghhBB1hjknHosx3WG7Vu+Pi2f1dLwrLi8vjzvuuINTp07x448/EuXkfZLBYGDu3Ll2r1PeffddkpOTWb16NcOGDbMbP3fuXOLi4mw/p6SkMGvWLLp3787WrVvR6ezTDhkZGWi1NffhjswrF3VeflYWp957D0tuLp4tWtDk8cdxDwtDcXHBPSzMrvCiqH3M+SoL/8jh+23Fu2conDZ/ykNdH2Jom6E1EpuoffwjI2k6efLl53jjxtTr3x+A5J9/JnHZMiq73rer1rXEOjQt6rVgzpA5JR5784Kb+d/O/5Ged/kFULp1Jzvy7ibdurPE49KtO9lpuJueHY+j18HxhHx+32eohKsRQgghRHUw58QT89NAYn8b5vAV89NAzDnxNRLXggULOHr0KM8//7zTBAiAm5sb06ZNIzQ01Lbt+PHjAAwYMMDpMY0aNbJ9f+rUKSwWC71793ZIgAD4+vri5eV1NZdxVWQmiKjTrGYzp99/H+P587jWr0+zp5/GxdeXgJ49azo0UQ5pWRY+/i2bmBQLigJ39nBjw8ELpGd5kmk5RKtQ+Hfff9d0mKKWcVZbxL1RI+IWLyZp9WrQaAi9885qi8dZ3ZD6HvUxWUzEZcYxa/Ms3v/rfW5udjMDmw/kyz1fcjLtpNN2u/EZ8aTlpTFr8yxOpp3k091vcU/EHP7Y68/KHXm0D3OhUX350y2EEEJUNVVVUS0lt7gvS35uIqrV6PzcViP5uYlo9f4VOreida/wzNdly5YB8PDDD1/Rcc2aNQNg/vz5PP3006XefuHYn3/+mWeeecYumVIbSIvcUkiL3NpNVVViPvuMtK1b0bi70/qVV3AvkoEUtduxeDP/W5dNVp6Kl5vCqJsM+Pqm8OQP/6WR8jin8v/L7JGP0tS/qUNdBSGcSf71V+K//RaAkFGjqm0ZnMliYsy3YwBsdUMAFo1exNoTa/l679ccOn+o1HNEtSuIdcVhJx2ugAjX9wjU9aNhgJYXR/vgopVlMUIIIURlKKn1qjU/l1NL29dgZCVrPuYQGp1HhY6tV68e+fn5ZGRk2G23WCykp9sv3fH09MTd3R0oqCPSqVMnMjMzCQsLo0+fPkRGRtKnTx+6dOnicDtPPfUUH374Ia6urnTv3p0ePXrQvXt3Bg4ciJ+f3xXHXZktciUJUgpJgtRuSatXk/DDD6DR0OLZZ/Hp0KGmQxJOpGVZyDJc/jWjqirRJ838vs+ACoTV1/L4bV50/6xViec49dypaohUXAuSf/mF+Estdv169sQQF4cxKQl9cDAhUVFV1irbZDHZ6oaoqmqrGwIFj/m9iXuZuWkmexL2VOj8M2/5gL/39iQrT2VwRzdG96rYCx8hhBBC2LvekiAuLi4EBwcTGxtrt/3gwYNERETYbZs5cyYvvPCC7eezZ8/y7rvvsmrVKuLjLy/niYiIYOHChXbJEFVV+frrr/n888/5+++/MV3q8KfX65kyZQozZsy4orogkgSpJpIEqb3Sd+zgzIcfAhA2YQKBt9xSwxEJZ8wWlRcWXSQzz/mvma7NXXjgZi90Wiv3Lb2PHXE7HMbMGTKHEe1GVHWo4hqS9PPPJHz/vf3GS4VUm06eXGWJkLKoqsqwRcM4knLEti3YK5j7O91vm1KqqiqL9i4iKTvJNiYiKIIV41ew/6yZj37NRgGeHeFN64aOnWWEEEIIcWVKenN9tcthjOmHift9dIn7Gw36Ab1/uwqd+2qWw5Q0EyQnJ4ft27cDsH//fp577jmHJEhRiYmJbN++na+++orVq1cTHBzMoUOHCAgIcBhrNBo5dOgQa9euZe7cuaSkpDBjxgymTZtW7rgrMwkiC4tFnZEeHU3iihUYExNR8/MBCBw8WBIgtZhOAwFeGrLyLBRPg/h7Kkwa5IkVC//+barTBEhEUATD2w6vnmDFNSN46FDO//Yb+UX/uF/qFpW0cmWNJUHMVjM6jc6u3S7Ag10ftM0aMVlM/Hr8V3zdfDlx4QRW1UqWMQuz1UzHpq70aatnyxEjn63L5v8Ge+Lmerm+ube7hgCvgp+Lz8ByNkYIIYQQJVMUBaWCsy0AFK3zN+pF91d0NsfVaN++PVu2bLElFAp5enpyy6X3Vc6KmRYXEhLCqFGjGDVqFOPGjeO7775jzZo1jB8/3mGsXq+nc+fOdO7cmVGjRtG2bVs+//zzK0qCVCZJgog6IT06uqAtZjFeraVtam2mKAojurvz/s/ZDvvuH+CJRbXwzC/PsOb4GjRoaOTbCF83X7s3iEWXFQhRXpbcXMeNqoohMbH6g7mktHa7zsbM2jyLz6I/w6pabR1vonq6s/Wokcw8lfdW2T+vfDwU3r7PD4C3lmU6nYFVOEZqigghhBBVS6v3R9HonRZHVTT6ChdFvVp33nknW7ZsYcGCBbz11luVcs6ePXvy3Xff2S2RKUnr1q3x9/cv19iqIkkQUSckrnBSLFBRSFq1Cv9u3ao/IFFuOo1tJQJQ8H14fS0tQqw8+dMU1p9cj4vGhXnD5tG/Wf9S3yCKa5+qquSmpgLgUb9+had66oODMcTFXX7gASgKbiEhlRFmhRV9PBdvt1t8zBM9n2DF4RWcyzjHV3u+4v+6/R9eeoUgXw1JF612xyiAv6cGBRVVBX8nM7AUIMBTg04mggghhBBVzsWzIeHDNmAxpjvs0+r9cfGsmcL/kyZN4uOPP+bdd9+la9euTtvkOquYsXHjRnr06GErlFrIarXy008/AdCuXcHynqSkJBITE+nUqZPDebZs2UJaWhqdO3eujMupEEmCiDrB6OzT2xr+VFeUbfcpEwt+z7Z7H6qqMKSLjsdXP86fZ/7EVevKJyM+oX+z/nbHlvQGUVzbzLm5vNegAQDTsrNx9fSs0HlCoqIKZo8VzcCparV1jKkMXq5ePNfnOab+NpWP/v6IqPZRBHoGcndvD4fZVSoQk2Lhsf9dLPF8KjCie8XXEAshhBDiyrh4NqyxZEdJPDw8+OWXXxg6dCijRo2if//+DB48mODgYDIzMzl69Cjff/89Wq2Wxo0b246bPXs227ZtY+jQoXTp0gVfX1+SkpJYvnw5u3fvZsCAAdxxxx0AxMXFERkZSWRkJIMGDaJp06aYTCb279/PN998g4uLCzNmzKipu0CSIKL2yzl5EtVicdxRCz7VFSX744CBJVtyUYGOTXSk5aicS7EQVl9hTvSj/BX7F246Nz6L+oybwm+q6XDFNcY/MhImTyZp5Ury4uLAakXj6opnixY1HdoVGdV+FIv3LuZA8gFmb5nN27e9TfswF8IDtcSkOPm9WAo3FwirV/4q7EIIIYS4NrVo0YK9e/eyYMECli9fzuzZs8nIyMDT05MWLVrw8MMPM3HiRNq0aWM75qWXXuKHH35g8+bNrFu3jrS0NDw9PWnbti2zZ8/miSeeQKMpmG7apk0bPvroI37//XeWLFlCcnIyZrOZkJAQRo4cybPPPut0lkh1ke4wpZDuMDUv88ABTv/3v1gvtVSyfap76d9mU6bg17VrzQZ5japoYUVVVVm5I481ewwA9G+vZ2wfD5bv3cfPO7VkuC7iQPoKPF08WTBqAd3CZDmTuMyUk8NMLy/g6maCFKVaLBx99VXyzp7Fr1s3mj311FWfszrtSdjD6G9Ho6Dw4/gf6RDcgYPnTHazQf5vsCdtGrqgKAXLXgCOxpv5dG2O3bl8PRQeusWLto2ks4wQQggB5es6ImqedIcR14X0nTs5+/HHqBYL3hERBPTuzflffsGQmIhbSAghUVGSAKkiZotarsKKxRMlFovKL3vy+OdsQfeekd3cubFFGofOn+argy9zJOcI5ICHiwfvDXlPEiDCjjknHkPa5SJZhrRDWI0eV71uVtFqCX/oIY7+5z9c3LmTi7t341ekj31t1zm0MyPbjWTl4ZW88ccbLB27lPZhLjQJ1HI2xUKTQC1dm7s6LHPp3MyVJoEGzqZYCA3QoKqQmG5l7uosbu/ixvBId7QaWRojhBBCiOuLJEFErZS6cSPnFi4EVcWve3eaPPooGp2Oer161XRo14WSWtsWLaxYWqIE4N6+HvS/wY3m7/V12JdrzuWxVY9x6rlTVXMB1zBzTnyZBbbKM6a2MefEE/PTQMy5ebZt8b+PRuemQdHoCR+2ARfPhhW+No8mTQgaMoTkn38m9quv8G7bFq1H9belq6jn+zzP2uNr2ZOwh5+O/sTwtsOJ6uHBkq05RPXwcFrnQ1EU25i7b/KkebCO77flsuWwkTW7DRxPyGd0T3e0TjrFSCtdIYQQQlyrJAkiaoX06GgSV6zAmJSE1sOD/IwMAOoPGEDYAw+gaOTFeHUqqbWtSkFy5Le9Bjz0oHdRwEkSpIGvhn7t9QC8c+s7TF071WHMnCFzqiT2a1lhoqCkVmvhwzYAlDnmapIJVcViTHcaM4BqNdpiLc+1lSQkKoqL0dEYk5OJX7qUxg88UCmxV4dg72Ae6/EYc7bO4Z0/3+GW5rfQLsyD18f6lXpcuzAXuzH39/ekbUMdizblcDIxn5k/Zjk9TlrpCiGEEOJaJUkQUePSo6PtujgUJkB8u3Yl7MEHpZNBDTDlqxyLNzvdt+eMmT1nnO8rNLZPwSfTJouJX4794rA/IiiC4W2HV0qs15PyJgqqOplQE7Jj16LzCCrz2kqLW+PqSuOJEzkxcyapGzYQ0LMnXq1bV1XIle7hrg+z9J+lxGXG8b+d/+Nfvf9VofNEttQT3kDH/9ZmcS7V6rBfWukKIYQQ4lomL3FEjUtcscK+jeUlpvPnJQFSA47Gm3nt+wx+2+v4ZvOmNq7c2tGNm9q4cmMTF5o10KIr0mxCUaBJoJb2YS7kW/N55pdn2Hx2MwoKzQKa8cagN4gIigDAbC09kSKuXPLf/yb573+XOsaUcQpzdmy5EiW1SfqhD0mJfvmqz+Pdrh31+vcHIObzzy8XXa4D9Do90/tPB2D+rvnEZcRV+FwNfLVMu9OXTk0dC6SqwIhul1vppmVZiEnJd/hKy3ZMoAghhBBC1HYyE0TUOGNiokMCBMCQmFgD0VwfnHV+yTNa2XzYRPTJgjeFfp4KY3t78Oseg6344oQBng6JqaJdKlQVRnR3R0Xlhd9e4Nfjv+KqdeXjER/Tv2l/FEVhbIexmK1mXLWu1XKttcHV1vHQuQeTm7SV9KMLSr0d08UjZcaSvP3p8gVdRkyVOVNEVVWyzvxY6hi3+l3JNySTnx171bfX8J57yNi7F2NiIkmrVxN6111Xfc7qMrjlYHqE9eDv2L95bs1zpOel88rAVyrUZlqnVXjsNi9e+jaD8xn2CY2lf+USn2ahS3NXZi4vu0iyEEIIIURdIUkQUaOyT5xAtTr5NFFRcAsJqf6ArgNlFTSFgra2UT3c8dBrcHPVlFp8sXiXinaNdLy24TVWHF6BVtHy/tD3GdBsgG28oijXXQLkaup4oGjRugZgMaaUeVv1OxXMEkjdO6PEMRpXX6ymjFLPc/HYV7gH9yJlxwuoVseZEpW5ZMZiyiB5+7PkxG8odVxg1/8AEPvbsBLHGC8exS3ghjJvU+fpSdiECZyZN4+kn3/Gv3t33MPCrizwGqIoCo90e4QdsTuIjo8GYNbmWbw56E0C3ANo6Htl/yeKojC2j4dd/R+tpqCLzLLteSzfnoebk6erLJkRQgghRF0lSRBRY1L/+IPYRYugMAlSuCTm0r8hUVE1G+A1qqTOLwA6LfxrmBetQi+/6yleWLG4oh0oRvZw572t77F432IUFN69/V0GtxxcNRdSR5RVxyMnYZPte+eDLFiMKWhc/fAI7k32uZ9LvC33oJ5lxtPw5sWoFiNxv5c8+yHrzDKyziwrcX956m8U52xWiSnjOKn73sWSlwSKC6CicTHTe3IgABqXgqSbotGj1fuXuUzn/N//xpx5ioCIp9Fo9aWO9Y+MJK1rVzJ27eLoq6+CqqIPDiYkKgr/yMhyX1dNeHD5g3Y/H0w+yMjFIwE4+exJu2TltphtvL7h9VJnixRPZD49zIvdp8z8ddTEqeR88pysGFIpmPUlSxaFEEIIUddIEkRUO6vZTNzXX5O6cSMAfpGR+Hbpwvk1azAkJuIWEkJIVBR+XbvWcKTXppI6vwA8eqt9AqS8ChMlH/39Ef/b+T8A3hj0BiPajbjqeK91KdEvlTmmXsep+LV+EIshlZy430ucVaLV+9u+L21MWckEj9CbMab9g8WQWs6rKF1ps2EAdB6hhPT9ny228Dvs9xddflPStRWUuLKSfvhTcuL/oF7HqejcGziMKnounxtvJGPXLtRLdUEMcXEFRZonT67ViZA5Q+bwzJpnnO7r8lEXWtZrSahPKA08G7D+1HrOpp8tdbZI0URmVA8PPN209G2vpW97N5LSLfx11MC6/UYsRSbtuWgh/kI+jevr8PHQOF1iB/atdssz5krGCSGEEEJUhCRBRLUyX7zI6fffJ+fkSVAUQu+6i6Bhw1AUhXo3XfmadlExrUN16LSQbyn4WVEgvL6WDuGORRLLY1vMNp755RlScwveNE/rN42xN46trHDrtPwyEgk6r/CCcdkxJY7xCO6NRqtH49mQ8GEbyqzTUZ4xpSVKGkS+Tr4hjbi1JXfwyTu/E71/exRFKbN2SGmzYQCCes6xLWMpbXaJSxnXb0w7xPno6ZgyjpP450NOz1F0KU/KunX2Oy/NREtaubJWJ0GGtx3Owt0LOZB8wLbNVeuK2WImw5DBrvhdEG9/TNHZIqeeO+VwzpJmfAX7axnV05OWoS7M++Vy4tRsgWXbDazYYaB9Yx0nEizkmUquGwKUuAyvaG2R0pbrSQ0SIYQQQlQGSYKIKpceHU3iihUYExJQVRWsVrQeHjR5/HF8b7yxpsO7Lq3bb7QlQOByQdMrndoenxFPzMUYnv/1eVsC5J6Ie7i91e2VGW6dVFjs8/yuV0odF9L7Q6D0WhdFuXg2LHMZSlljykomFCYuSpO65w2yzq7Ct+V9pES/VGrNE4vhQqnn0rh4lrq/eOwlXZuLZ0PcA7uS+NcU8pK2Oh1TdCmPMSnJyQC11hdlLuysFBEUwZgOY1j6z1IAvh7zNXEZcRxPPc4vx35hwynHOivv3f5ehW7zhsaXl8yE1dfSp60r24+ZOHPewj9n80s8Tq9T2HvKSL4VXHQKOCzCA60CG/bnoSgKqqqidTLZQ2qQCCGEEKKySBJEVKn06OiC6eXFhIweLQmQGpKSYeGXXXkA1PfWkJpltbW1La8MQwa/n/idqWunOuxbcmAJSw4scfpp87XI2SwIizGd9MOfkZfs/I14bVCeZEqpNHqMaf9wfsfzJQ5RrUbifh9Nfm71JRW0bgHU7ziV2N/Kvu/1wcEY4uLsu1PVgaLMrlpXlo5biovGxaHjUtsGbWnboC3D2w4nanGU3WwRgLUn1nJ7q9txc3G7otssumTmrp4etAtzYUCEO/Fp+fx11MTWwwZyndQOScm0Mn99bqnnTs9RWf63odQxKjC4k5td215ZMiOEEEKIipAkiKhSiStWOG5UFC5s3EiDW26p/oCuc6qq8u2WXMwWaNNQx22d3fh+a26JnV/gcmHF5/o8R5YpizXH1rD17Fbbp9HOzBkyp6ouoVYpq9YFig6/Ng+RcezLq67jUd20ev9SY2o0eBlZMau5eOwrcNJBplBBAsT5DICaFhIVVZCkLSzKDHWmKHPRDkvOOi4Vny0yf+d8YjNi+f3k74z/YTzzo+bj735ljytnS2YaBugY3UvHyG5uvPp9pl2rXVcdNKqnxUWroNUWzPg4lZRvlyzx1Cvc0Fhn9/tHVVUOnssnx2j/mFnwew67Tpro0cqVr//MJUuWzAghhBCiAiQJIqpUXZ1ufq3ac9rMwXNmdBq4t68nwf5aXh/rvBBqfEY8aXlpvLjuRWIzYnls1WOoRd7Itqrfittb3c6vx3/leOpx2/aIoAiGty25lsS1pKxaF8G95uIdPhS/VvdVSh2P6lSeJTNuATfgEdyHhI33l3ie+p1eQh9wA/Eb7qnKcCvEPzISJk8maeVK8uLjwWJB4+6OT0RETYd21ZzNFtkes50nfnqCvQl7Gf3taBbeuZAwv8ppDeyi0zi02n3sNi9uaGz/++XgOZPdmIcHeTqMcTauga+G8xlW9pw2s+e0GWc5DlkyI4QQQpRt06ZNDBgwAIAnnniCDz/80GHM+fPnadSoEWazmX79+rFp0ya7/fv27eODDz5g06ZNJCQkoCgKTZo04eabb+aRRx4hothrKYvFwjfffMO3337Lnj17uHjxIl5eXkRERDBixAgmTZqEt7d3lV1zcZIEEVVK5+OD+UKxegB1YLr5tchgUlmyNQeAWzu5EeyvLXV83/l97X4umgD57YHfaFm/JSaLiT9O/eFQm6Bwav71zsW7ScG/lVDHoyaUJ6ayZqm4B3Uvc1ZJTcx0KeQfGYl/ZCRqfj6Hnn8eU2oqqZs20eDWW2sspspSfLZIrya9WDp2KROXT+RM+hnu+vYuFoxaQERw5SR9irfadbbErjxjnI2bfpcP8RcsbDtq5O/jJrKdLIWRtr1CCCFE+bm5ufHtt98ye/Zs9Hq93b6vv/4aVVXR6RzTBW+//TbTp0+nXr16jBs3jnbt2qEoCkePHmX58uV88sknxMTE0KhRIwDS0tIYMWIEW7duJTIyksmTJ9OwYUMyMjLYvn0706ZN47fffmNd8YL1VUiSIKLKqPn5qOZiSyYuTTuvC9PNrzWro/O4mKMS6KNhSBf3UscmZCbQyKcRcZlxDvvmDJlDy/otgdJrEwhRqDyzSipTaUkXFJ3TpIui0xE0bBixCxeS/Msv1L/5ZjQuFeuWVJu1rN+SZfcu46HlD3Ek5Qjjvh/HB8M+wEXrwusbXueVga9wU3jFOnUVb7XrLBlRnjEljWtUX8fdvXXc2dODfWdMLNqUQ16RpTUKsPukCXdXDc2CtKRnW6VuiBBCiBpVm2tYRUVF8d1337Fq1SrGjBljt2/hwoUMGTKEDRvsi6wvXryYadOm0bdvX1atWoWfn5/d/nfeeYd33nmnoBnGJWPGjGHr1q3MmTOHf/3rX3bjn376ac6dO8cXX3xRuRdXBkUtGqGwk5mZia+vLxkZGfj4+NR0OHVOyvr1xH71FRoPD1wDAjAmJ+MWEkJIVBR+XbvWdHjXldjUfN78IROrClOGOk5RL2pH7A6eXP0kaXlpaBUtFvVyG5mIoAhWjF8hn7ReYkg7WGpXl7DbfrK1fr1WlVYXpWg72pqIq2jSJevcGi4e/gS0bjQZut5pTFazmUPPPos5PZ3GDz5I/Ztvrs6Qq1WWMYsnVj/BtphtaNAQ4hNCfGY8NwTdwJuD3iTAPYCGvrVrZlJxxZfMFBXspyE9x4rRSekiqRsihBCiKIPBwJkzZ2jatClubldWOLw0ZovKC4su1rq274XLYWbOnMkPP/xAgwYN+PXXX237d+7cSffu3VmxYgXjx4+na9eubNq0CZPJRNOmTcnIyOD06dM0aNCgzNv65ZdfGDp0KKNHj2bp0qVXFXd5/p/K+/5dZoKIKmExGGxFURuOHk2gFEGtMVZV5Zs/c7Gq0KW5S4kJEFVV+WrvV8zYOAOLaqFtYFusqhVXrassdSmBaim9o8X1oLJneVgtFs5t2QJA4z590GhLX7ZVWlxFb1vv3w5DSjSGlF2cj36Z0H6fOyTzNC4uBN1xB3GLF5P000/U69sXxck00GuBt96bBaMW0HZuW6xYic+MB+Bg8kFGLh4JUOs7PBVdMhMeqOWunu5sO2piz2kTSRetTo+RuiFCCCHKS1VVTCV3gS/X8X6eGrLyLHbl4RXAz0ODxaJidf7nqkyuOirlQ8kHH3yQKVOmEBcXZ1u+8sUXX9CgQQOGDh1qN/avv/4iISGB8ePHlysBAvDDDz8A8Mgjj1x1rJXp2nx1J2rc+TVryM/MRB8URP3+/Ws6nOva1iNGTiXno3eBu2/ydDrGYDbw0u8vseJwQeJqeNvhzBg8A61GK0tdSpF55scS99V0rYvqVJn1TPINBr66VKxrWnY2rp7OH7NXSlE0BHV7m3O/DiE3YSPZMT/h3cSxgG/9/v1JWrUKU2oqaX//Tb3evSvl9msjV60rs4fM5tk1zzrsm9xzcg1EdGWKLpkZ1cODNo1caNPIlbFGK9EnTfy+z0Byhv2rS6kbIoQQorxM+fDkfMcPea6WCpxLtfDUgosVPseHk/zRV8Kq3XHjxvHcc8+xaNEipk+fTl5eHkuWLOHhhx92qAdy4MABADp16lTu81fkmOogn4WISmfOyCD50pSq0NGjr9lPUuuCzFwry7fnATCymwf+xdYebovZxs0LbuaOr+5gxeEVaBQN0/tPZ86QObi7uOOqdbW9WXDWhvN6Zkw/TOapgtkx9Tv/h7DbfrL7qqllIKJkrr7NCbjhSQBSdr+GxZDmMEaj19NgyBAAklavRq3oRzR1xIi2I4gIciyMOm/7PO7+7m7+OPUHVrXgPtgWs41bv7iVbTHbqjvMEhW27W1XpMCqh15Dv/ZuvDHOl1B/+995ehdoGFCx2UVCCCHEtSYgIIARI0bw5ZdfAvDjjz+SkZHBxIkTHcZmZmYCXFGZiIocUx3k3amodEmrVmE1GPBo1gy/bt1qOpzrSvHiS6t25pJrVAnx1zAg4nLV58L2t/9Z/x9iLsYABdPjXxv4GiPajaj2uOsa1Woheed0UC14NR6Cf5sHajokUU7+bR8hK+YXTBnHSNnzJsG95jiMCRw4kOSff8aYmMjF6Gj8u3evgUirh9laUDSjsMPT13u+JiU3hSxDFrvid7FrxS6a+DVheNvhbDi1gZNpJ5m1eVadqBuiKAqjb7Jv22s0w1vLMnn0Vi9ahFx7hW+FEEJUHlddwYyLq3U41szHv13+W/T4bV52yfuKcK3Ed/EPPvggS5cuZevWrXzxxRd069aNdu3aOYwrTGQUJjbKo+gxAQEBlRNwJZAkiKhUxuRkUv74A4CGd98tU46rkdmi8tayTKfFlzJzVawqFH7+Wbz9LRQUSnxmzTOSBCmHjJPfYLywH42LN4GdX6npcMQVULSuBHWfSey6O8k6uwLvJiPwDO1nN0br7k6DwYNJXLGCxFWr8IuMRNFcmxMnS+rwlJ6XzsLdC/lu/3ecvXiWedvn2Y6pq3VDGgZosaoqielW3luVxeheHtwcoZe/U0IIIZxSFKVSlpx0bGrf9r1jU5da9bdn8ODBNGrUiNdee42NGzfyySefOB0XEVEwc3Tv3r3lPndERAR79uxh7969DBw4sFLirQzX5qs6UWMSli0DiwWfDh3wdpJBFFVHp4EALw3OfqUG+tgXApwzxPHT79K2X2vMOfEY0g46fJlz4ss8Nj83mQv73gWg3o3Po/MIqupwRSVzq98Jv9YPAJD89/PknY92eBwEDh6Mxs0NQ2wsGfv21Wi8Vc3ZsrcgryBe6PcCWx/Zyh2t73B63Du3vVOdYVZIYd2QEH8NY27yYPpdvkS2cMVihSVbc/nkt2xOJpqJScm3+0rLvraXQQkhhKg+Rf8WldYevqZoNBruv/9+1q9fj5ubG/fcc4/Tcb169SI0NJSVK1eSkpJSrnPfddddAMyfP7/S4q0MMhNEVJrcM2dI//tvUBRCi/WaFlevrD7jiqIwuJMbn63LcRhTvBBgI99GDmMigiIY3taxUOS15mpbuqbsfg1rfjZu9Tri22JcVYYqqpBPi3FcPLYQiyGFuPX2v68KHweBt9xC8s8/k7RqFb6dOtW6Fy3VwVvvzftD3yfmYgwHkw/a7ZuzZQ4Gs4HREaPR6/QlnKHmFdYNKTRpkCdNGmhZ9lcee8+Y2XvGsY+utNEVQghRmYr/LaptHnnkEVxcXGjWrBm+vr5Ox7i6ujJjxgweeOAB7r77blauXOlQ68NkMvHOO+/wwAMPEBYWxh133MGAAQP4/vvvuemmm3jqqacczhsTE8MXX3zBa6+9ViXX5owkQUSlUFWV+O+/ByCgVy88wsNrOKJrS2lLXbzcFG6O0HPoXD6nku37eClAeKCW9sXWHc6PLsjG+rv780zvZ66r9rcWY7rTBAiAajViMaaXmATJif+D7NhfQdHSoNtbKBopsFhXldbeuPBx0OC22zi/bh25p0+TdfAgPhGOBUSvB2arGQWFiKAI7rzhTuZHzyc5O5nknGT+s+E/fLrzUx7v/jh3RdxFdFw0r294nVcGvsJN4TfVdOhOKYrC4I7uNK6vZe5P2ViL/VqVNrpCCCGuN40bN+bVV18tc9yECROIi4vj5ZdfpkWLFowbN85WP+To0aMsX76cuLg4W2FVRVFYunQpw4cPZ/LkySxevJjhw4cTGhpKZmYmf/31FytXrmTApc6A1UWSIKJSZB08SNahQyg6HSF33lnT4VxzCpe6FO8zDpBtUFkdffkNXaCPhpTMgqncztpBnkk/w/qT6wH4dsy3tApsJe1vy8Gan8v5XQX1P/zbPITeX5Z7XetcfH0JHDCA82vXkrRq1XWbBCleN2R8x/Fkm7JZdXgVH+/4mMSsRF5e/zLv//U+rlpXErIS6kTx1DaNXHngZk++2GA/e04FhnWTNrpCCCGEMy+++CK33347H3zwAatXr+bTTz9FURSaNm3KiBEjePTRR2nY8PLf/vr167N582YWL17Mt99+y3//+18uXryIt7c3ERERzJo1i4ceeqhar0GSIOKqqVarbRZI4C23oA8MrOGIrj2KojCiu7tdl4NCGg20b+RChyYudGjiir+nwoxlmbbiS8VngXyx6wtUVAY0G0CrwFa280sCpEDWmR9x8WyEVu+HOScei7GgP/zFowvJz4lH6xaId7PRNRylqC4Nhgzh/O+/k33sGHsfeAB9aCghUVH4R0bWdGjVqujvB0VR8NZ7M77TeEZHjGbJP0t4/Y/XSc1NtY0pWjx131P78NZ72/Zti9lWrtki5R13NXq0cmXDPwZiUix22xdtzKZ3Wzf6tNNTz1tb5nJEKHvJYmWOEUIIISqqf//+qKrj3xlnsrMd33sAdO7cmYULF5b7NnU6HQ888AAPPPBAuY+pSpIEEVclPTqauG++wXzhAigK7mFhNR3SNatVqA69S0GLx0INfDW8PNoHN1f7F8ZRPTxYsjXHofhSak4qyw8tB2BS5KRqibuuuXhsIRknv8Wz4S1kx60Dq329AIshhdhfh5ZZO0RcG3JOnQLrpZlVFguGuDjOzJsHkydfd4kQZ/Q6PRM6T8DTxZOpa6c6HdPpg040D2hOi3otaOzXmA2nNnAq7VSJs0UKW3jP2jyrylvyKorCyGIJZndXyMiFX3YbWLPbQLswHaeTLeSZHF8wFtYOAUpcsljZY6ROiRBCCHF1JAkiKiw9OrrgzUAhVSVm/nw07u7y5qCSqarKoo25dgkQgLF9PBwSIFBy8aXF+xZjzDcSERRBt0bdqija2s2cHVfqfhfvppizzpB97pcSx5RVO0RUnIu7O48dPGj7vqYlrlhhv+HSJydJK1fK77ki7rzhThbvW8yB5AO2bb56X7xcvYjPiudk2klOpp20O6bobBGlSF8rtdiiv6puyVu0jW6TQC3/jvJm31kzWw4bORKXz6HY/BKPddUqRJ8oqDHkqlPAYcFi5YyROiVCCCFE5ZEkiKgwhzcHAIoibw6qwOroPHacMKFRCmp+JGdYnS51KU2eOY/F+xYDBbNArsf17hZTBql7Z5S4X9HoCR2wCEteMhcOzCMvaXM1RicAFI2GBu3bV/ntaPX+KBq98yK5Gle0en8AjElJTo/Pi43FkJCAW2hoVYZZZ5gvzZiKCIpgTIcxtmLLS8ctJdOQyYGkAyw9sJR1J9c5Pb544sOZBzo/gFW1olEqNxNQ2LqwcPaci05DZAs9kS30JF+0sOWIkc2HDOSZHI9NzbKy8I/cUs9fGWNUoGcbvd3vbVl+I4QQQlSMopZ3QdB1KDMzE19fXzIyMhza/wjYO3EiqtmxtaDi4kKnL76ogYiuTVuPGPlqY0HhvgkDPAnw0rBkaw739Pak3RUkQRbvXcx/NvyHMN8w1j+0Hp3m+sqBqlYLCX9OJDdxM1q3BgT1nI1W72c3Rqv3t83uMKQdJPa3YSWeL+y2n3ALuKEqQxZVrGjNF1VVSYl+CWPaP3iEDqRh/wUAHJ4+HUNcnG0GiB2Nhvr9+xMSFYWLn181Rl47mSwmW/FUVVUdii2rqkrU4ii72SJtA9vyxZ1f2L25V1WVicsnciTliMNtNAtoxqTISYxoOwK9Tl8tdUMATPlWXl2SaSs6DeDmAs2CdLbYVVXldHI+hiJ/FitrTKEmDbT0a+9Gx6Yu/Oe7jDKX1ryw6OJVj5HlN0KIa53BYODMmTM0bdoUNze3mg5HlKA8/0/lff9+fb0LEpVKHxyMITbWfqOi4BYSUjMBXYMOx5pZ/GdBAuSOLm70bqsHuOI+4xarhc93fw7AxC4Tr7sECEDqvrfJTdyMonWj4YCF0t1F4OLZ0G5JU1D3mZz7dSi5CRvIS9mNe2AXQqKiCpb9KUpBIuTSvx7NmpF7+jSpf/xB2rZt+Nx4I4aEBIzJyeiDg6V4qpNiyyXNFvFz97Mba7KY0Gl0tnHf7vuW1NxU8sx5nE47zbS103hv83vc0eYOouOiq7xuCICrTsO4vh52tUMeudWLGxrbX+PBc6YqGdMqVMeppHzOnrdw9nwO328FF51jcqL4shlnXcUqMkYIIYS4llx/74REpQm85RZii1YFvvTmICQqquaCuobEXcjn07XZWKzQraUrI7pVvD7CuhPrOHfxHH5uftx1w12VGGXdkHl6ORePFnyyH9TzPUmACKf0/u3waT6GzFPfk7LnDcIG/1iQyJg8maSVKzEkJuIWEkJIVBR+XbuSdfQo8d99R+7p01zcudN2Hime6lzxVrslteYuaZzJYmLJ/iUs3L2QpOwkFu1dZDumquuGgGPtEGfLEatqzHMjvMnKU/nrmJHNh4ykZFoxmB1nb6iAh16xzR700CsOC43KO6Z4e3UhhBDiWiFJEFFhGn3BrATF1RVU1e7Ngbg6F3OsfPBLNnkmlVahOh642bPCL0ZVVWV+9HwAxnccj4erR2WGWq2KLl8oqugyluJjjBePcn7HNAAC2j+Jd+M7ynVbpdWMUDR6W80IcW2p1+FZsmJ+xnhhP1kxP+HTZAT+kZFOkxnebdrQ+tVXOfivfxV0yCp0aelM4ooVkgQppqzZIqWNc9W68nDkw9zf+X5e2/AaS/5Z4nDcG7e8UflBF4mjpM5b1THGx0Phtk7uDO7oxtG4fDYdzGPvGceirYfjSi7kWt4xehfIM6rkW1R0WkXqhgghrgtSJaJ2q8z/H6kJUgqpCVK6cwsXkvrHHzS4/XYajRtX0+HUaUVfYJrMKl9tyiH5opVAH4UX7/LF063iLzJ3xu5k7Pdj0ev0bJ60mfqe9Ssr7Gplzokn5qeBJSYlwodtAChxDGgIH/4nrl6Nrug2y0q6iGtP2qGPuLD/PXQeoYQPXY9GV/osrJLqIwGEP/ooAT17omjkjWJlclZfBMDTxZOJXScysctEfNyu/b/bO44bWbA+x/bzTW1cCfLT2o1Jvmhh21HTFY0p5Ouh0KuNK1sOm8h2kgSRuiFCiGuBxWLh+PHjNGjQgHr16tV0OKIEqamppKam0rJlS7RardMxUhNEVLnsY8cA8GrVqoYjqdvMFpW3lmU6LUyXZwJXl6t7cblgV8EykFHtR9XZBAiAxZheQnLjcsvawu+ds2I1XQTKnwQpXjNCXB/8Wj9Exolvyc9N4OLRBQTc8FSp4/XBwSUWT4359FOSVq0iZORI/Hv04OLu3SSuWIExKanE2iHp0dFljrne2dUXiRjDgl0LSM5OJsecwwfbP+DLPV8ysctEHuzyIP8k/VMtxVNrQreWrqzfb7Atm5kwwHHWoKqqxF+wlHtMo3paOoTr2HrEREauyq97nP9OlbohQohrhVarxc/Pj/PnzwPg4eF8lp6ofqqqkp+fT2ZmJpmZmfj5+ZWYALkSMhOkFDITpGT52dn889hjAER89BEucv9UmKqqzFiWSUyKxWFddpNALdPv8qnwL+KTF05y68JbUVD4feLvNA1oevUB15CyurW4eDcDwJx1usQx0tFFlFfW2dUk/TUFRedBk2Eb0bk3KHFsenS00+Kp/r16kfnPP1iyCwpcuvj7Y05PdxjXtEjtkJLO1VTqizgo3o3GaDGy8dRG3v/rfU5cOAEUzAzxdPXkfM55bgi6oUqLp9aUw7HmMjuGVWRMvkVl7xkTfx40cizB+fKZKUMdi7oKIURdpKoqSUlJXLx4saZDEU5otVoaNGiAr69vqe+Lyvv+XZIgpZAkSMky9u7l1Jw56ENCaD9rVk2HU+cdjDHx/i/ZDtuv9gXmtLXTWHpgKYNaDOLTkZ9eTYg1rqwkSHlIEkSUl6qqxK27E8OFvfg0G0NQj3dKHZ8eHe20eKolL4/za9dy/tdfseTmOj1W6+GBb6dOQMHvVodxioJ7WBht33qrUq7tWmdVrfx67Fcm/zy5xDFVVTz1WhV/IZ/Zq7PIKjJjsZ63hpnjS38xKoQQdY3FYsFcwhJXUTN0Oh1arbZcf29kOYyoUtnHjwOyFKYyWFWVPaft12IrCoTXd941oLzOZ59n5eGVAEyKnHQ1IdYJ9bu8CkDq7ldrNA5xdSxmM7s/+wyALv/3f2hdKv4cuBqKolC/y8vErRtF5umluAf1wtW3uW1/8bowJRVP1bq7EzJyJA0GD2b/Y4+B1eowxpKbS9q2bSUHo6oYEhOv7oKuIxpFwx1t7sBkMfHcr8857J8zZE4NRFW3NaynY+JAT7u2vReyrCz7K487e7qj0UgiRAhxbdBqtZWy3ELUbpIEERViqwfSunUNR1K3Wa0qX27MYfsx+ySIql5de8JtMduY8tMUTBYTnUM706Vhl8oIt0aVNWnNPbDuX6MAi8nEr08+CUDHBx6osSQIcGkJjAawkrz9abt9hcV4y1szRuvhgVvDhk5rh+j8/AgaMgSA5DVryC8+FVdRcAsJqdhFXMdGthvJV3u+siueqlW0tK4vf7cqomjbXl8PhYxclXX7DSRdtDBpkBdurpIIEUIIUTdIOStxxawmE7mnC+ouyEyQisu3qCxYX5AA0Sjw0C0eNAksyDw3CazYLJD4jHgOJB1g5p8zSTcUFAod3HIw8RnxlRp7TTCk7i5xX2HL2sK2tqWNEaK8CortOs7cAPtivOUVEhV1uc4H2P5tPGECQbffTtDttxN2//1ObkwtOFZckaLFU1/s/yLuOncsqoXxS8dzPPV4DUdX9xS27Q3x1zBxoBeTBnmi08I/MWbe/jGTC1mWmg5RCCGEKBepCVIKqQniXPaxYxx/8010vr5EfPCBrAeuALNF5bN12ew7Y0argf8b5EXn5q7lKl5XmubvNS9xX11eA28xZRLz80AshlR8W96HT/MxdvuLLk2QtrZ1myknh5leXgBMy87G1dOzxmIpqw5NRWrMlFQ7pPiYxBUrbLNG/Lp1o9lTpXeoEc4VLZ56Me8i438Yz5HzR6jnUY9v7/6WFvVa1HSIddrp5Hw+WpNFZp6Kt7vCff08CPB2nEbu7a4hwKvgc7eiLeFLGiOEEEJURJ2uCfLdd9/x3nvvcfjwYTw9PRk0aBBvv/024eHh5TreYrEwf/58Fi5cyOHDh1FVlaZNmzJmzBhefvnlKo7+2le0Na4kQK6cKV/lk9+yOXjOjE4Lj9/mRUR4QfHTdmEuvD7Wr0LnNeYbGdlupK0OSFF1fQ38hX/mYDGk4uLTjPqdX0SjdT7bA6StrajdSqod4mxMYaeYrIMHseTloXV3r6Yorx2u2suFpf3c/Vg8ejH3/XAfh88fZvzS8Xx797c0C2hWgxHWbc2CdLx4lw8frMkm7oKFj3/LcTrOx0Ph7fv8AEpsCV84xkUrryuEEEJUrVqXcv/www8ZN24c7u7uzJ07l6effprff/+dXr16kZCQUObxZrOZkSNH8tRTT9GqVSveffdd5syZw7Bhwzh37lw1XMG1z1YUVeqBlEtaloWYlHxiUvI5kWhm1opMDp4z46KFyXd42xIgFWXMN/L13q+5ecHNThMgEUERDG87/KpuoyYZ0g6SceJrABp0fb3UBIgQ1Sl13zsYM05gzonHkHbQ4cucc3XL0Py6dEEfGoolN5fUP/6opKivb37ufiwavYg2gW1IyUnh3u/v5UzamZoOq04L8NYydZQPNzZx/rmaAvh5aMjPV8nPV/Hz1FA8zaEAAZ4adLXuVakQQohrUa2aCXLhwgWmTZtG586d2bRpEzpdQXi33XYb3bp145VXXmHBggWlnuPNN99kzZo1rFmzhltvvbU6wr6uqFYrOSdOAFIPpDzMFrXET71cdAotQq78KbgtZhuvb3idaf2nkZCZwEc7PiIpKwmAYK9gtBot/u7+3N3hbpb+s7QgDqvZ7hPRukJVraREvwyqFa/wYXgE31TTIQlhk5e0lXO/DKaweGpxV1o81fF4DcF33EHM/Pkk//orgYMGoXGte8/j2sbf3Z9FoxfZaoPcu/Renu39LJ/t/IxXBr7CTeHye+ZKubkoPH67N5+ty2b3KfvWkipwLtXC5M8vlni8ytUVAxdCCCGuRK1KgqxatYrs7GwmT55sS4AAdO3alb59+7J06VI+/vhjXEt4EZiTk8N///tfhg0bxq233oqqqmRnZ+Pt7V1dl3DNM8THY8nNRePmhnvjxjUdTq2n00CAl4asPAvF0yANfK7sU6/4jHjS8tJ45893OJl2kkdXPmor/BfsFczjPR7nrhvuQlEU2xr4sR3G1tkECEDmqe8xXNiHRudFYKcXazoccZ0pLLSrWo2OOzUuuAf1Ii/xT8oqnno1y7MCevUi8ccfMV24wIUtWwgcOLDC5xKX1fOox9ejv+bu7+7m7MWzTF83nXxrPrM2z+LNQW8S4B5AQ19ZVnclNIrCI4O9mLY4gwtZzp8TJQnx11xVS3ghhBDiStSqJMjOnTsB6NWrl8O+Xr168eeff3L06FE6dOjg9PitW7eSmZlJt27deO6555g/fz6ZmZn4+/tz77338s477+Dh4VHi7RuNRozGyy92MzMzr/KKrj2F9UA8W7RAkR7aZVIUhRHd3Xn/52yHfVf6qVff+X3tfi5MgAD88fAf6HWOy0QURamzCZB8wwVS970DQECHZ9B5BNVwROJ64+LZkPBhG0ottJt1djVJf02pshgUnY4Gd9xB3KJFJP/yC/X795ffvZWkvmd9zl48C0C+NR+Ag8kHGbl4JFC3i0nXFEVRGN/Pw+5v3pO3ezkU+j4ca+bDXy+PScmwsv+smY5N6+bfKyGEEHVLrVp9GR9fsH66UaNGDvsKt8XFxZV4/NGjRwH473//yzfffMObb77JsmXLuOOOO/jwww8ZPnw4pTXDmTlzJr6+vravsLCwq7mca5KtHogshSk3g8n+MacoFWuBW1Jx0zlD5jhNgNR1F/bNwmrKwNWvLX6t7qvpcMR1ysWzIW4BNzh8Fc7ucPEpvaimmp971THU79sXnbc3ppQU0v/++6rPJy4r6ffqkz2erOZIrh3tw1zs2r13aOKCi06x++rQ5PIYNxfIt8LHv2azfr+hJkMXQghxnahVSZDc3IIXi3q94xs6Nzc3uzHOZGVlAZCWlsaGDRt46qmnuPPOO/n6668ZP348GzZsYN26dSUeP23aNDIyMmxfsbGxV3M51yRJglyZrDwr326xf8yqasXWPjf2c1x+VNeLnhZVtLhkxqmlZJ4uqGcScMMUFE2tmrQmRLklbH6Eiye+QbWay1VA1dkYU84J6g/uBkDSTz+hWq9sqYEo2fC2w4kIinDY/uHfH3L/D/ezO353DURVtymKQlQPD0L8NUT18HD6t67omP8b7EXfdnpU4PttuXy3JQerteQPrIQQQoirVaveWRQuVTEajbgXawWYl5dnN8aZwmO6d+9Ou3bt7PZNnDiRxYsXs3HjxhILpur1eqcJGFHAlJqK+cIF0GjwaN68psOpE77dnEtWnkqovwYXnUJMiqVCs0DyzHk8t+Y5APzc/Hi2z7N1vuhpUeaceGJ+Gui09kLyX1Nwq1fx4pKibnH19OQ/pczYq2uspoukRL9E+pHPyM9JADXfYUxhAVWgxOeBonFF69cNQ3w8Gfv24de5c5XHfj0oXFYYERTBmA5j+GbvN5zPOU+GIYNtMdvYFrONPk368PRNT9MxpKOtMHVpBVSrc0xtVZ5270XH3NDYhQa+GpZtz+OPA0YS0y0Mj3THRWefQPF21xDgVas+vxNCCFEH1aokSMOGBW9y4uLiaNmypd2+0pbKFCrcFxIS4rCvcFtaWlqlxHo9KpwF4tGkCdpLM3NEyXafMrHrlAmNAhMHepFjVFmyNafET8ZKM3vrbM5ePEugRyC/PvAr/h7+db7oaVEWY7rz4pNUTnFJIapKacVTFY0r/u0fJ+P4IvKzS27RXvgYL/ze+RgTAb07k/LzHpJWr8a3UyfppFEJXLWuLB231KGYdHJWMh/v+JjlB5ez5ewWtpzdQpfQLlw0XORU2imnBVQLi1fP2jyLk2knq3TMtUZRFG7t5E59Hy0Lfs/mSFw+R+KyHMb5eCi8fZ8fLlp57AshhKi4WpUEiYyM5H//+x9//fWXQxLkr7/+wsvLizZt2pR4fPfu3QGcLmM5d67gBWhQkBRXrChZClN+WXlWvtmcA8Dtnd0Ib1DwVCvrkzFndsTu4MvdXwIw87aZ+Hv4A3W76KkQ14ryFE/1azORlN1vknVpiZczsb8NK/O2/Hv2JHXtAXJPnSL7yBG8i814FBVT9Pdo4e/VML8wZt46k0e7P8pH2z9i+aHl7E64vDSmaAHVklTFmGu9WGuX5q74enjx7spsiq+IUYAAzyvrqiaEEEI4U6v+lIwYMQIPDw/mzZtHfv7lKcO7du1i8+bNjBkzxtYeNzExkaNHj9rVCGnSpAl9+/YlOjqav4sUj1NVlY8++giAIUOGVNPVXHtsnWEkCVKmby4tg2kYoGVoV/eyDyhBjimHqb9NRUVlTMQYBjQbUIlRCiEqQ1nFU7Uu3pVS3Ffn5UX9/v0BSFq9+qrPJ8oW7hfOrNtnMa3ftJoOhb5N+pKSk1LTYVS5FiGu3N/fcemzSsXqaQkhhBDF1aqZIPXr12fGjBk8/fTT9O/fn/vuu4/U1FTmzp1LUFAQr7/+um3stGnT+Oqrr9i4cSP9L70oBPjggw/o3bs3t956K0899RQhISGsXr2adevWMXHiRHr27FkDV1b35efkYLjUmUdmgpQu+qSR3ZeWwTw40BPdVUzbffvPt4nNiCXUO5Tp/adXYpS1i6pKoUdxfQsdsAiAhI33lzquwZAhpPzxB1mHDpFz+jSezUrvTiMqx0NdH+Lnoz9zIPmAbVvbwLYsvHOh7U25qqo8uPxBjqQcqfQxAJvPbqb//P7c2/Fe/i/y/6jvWb/Krrem9WqjZ8MBI7GpFrvtGw8YcNEqtArVkZ5tJcvgWD9I6oYIIYQoS61KggBMmTKF+vXrM3v2bJ5++mk8PDwYNGgQM2fOtNUMKU2HDh3Yvn07L7/8Mh9//DE5OTm0aNGCuXPnMnny5Gq4gmtTzokTAOiDg3Hx9a3haGqvzFwr324umJ00pIsb4YEVf4ptPrOZb/d/C8A7t7+Dt967UmKsjTJOfFPTIQhRo7R6/3KN09evT0DPnqRt3UrSTz/RfMqUKo5MgGMB1cLC1L7uvrblNCaLCZ1GVyVjsk3Z+Oh92J+0n893fc63+77lvk73MSlyEkdSjlxzhVgVRWFUD3fe/znbbvs/Mfn8E5NFWD0NKVlWDCbHY6VuiBBCiLIoqnoNleGvZJmZmfj6+pKRkYGPj09Nh1Oj4r//nuSff6Ze376ET5pU0+HUSqqq8unabPacNtOonpYX7/Kp8CyQTEMmt395O0nZSdzX6T5eHfhq5QZbi2SeXk7y38+VuL+wc4YURhV1WWkdkMrXHeby8yAvPp4jL7xQsF2nQx8SQkhUFP6RkVV7Edc5k8VkK6CqqqrTwtRVOcZF48KfZ/7k/b/e55+kfwBw07nh5epFam4qYb5h3NvxXjxdPannUQ+AC7kXyDHl8M2+b4jNiC3XmBsa3MCbg2u+EKuqqsxYlsnZS13VHhzoyR8HjGw/ZsTk2GQJKKgbEh6oZfpdPrJsRgghrkPlff8uSZBSSBLksmNvvEHO8eOET5pEvb59azqcWiEty2I3FffwORM/7jCgUeDFu3xofBWzQJ7/9Xl+PPQj4X7h/Hz/z3i4ltwaui7LO7+TuD/Gg9WMb8v78Gk+xmFMYXFJIeo6c058qQVUi4/Jz00icfMjgJWQvgvwajQQgPToaM7Mm3f5BIoCqkrTyZMlEXIdUFWVTac38fCKh6v8tmq6EOvhWDNLtuZwT29P2l1qLZ9tsLL5kJF1+/LIcdJMacpQL25oLEXDhRDielTe9++1bjmMqH2sJhO5p08DUhS1kNmi8tayTDLzHHOILloICdBW+Ny/n/ydHw/9iILCrNtnXbMJEFPWWRK2PApWM15htxPY9VUURdZxi2uXi2fDMhN6dmMCbsC76UiyzvxI5umltiRI4ooV9gepKigKSStXShLkOqAoCgOaD2D2kNk8u+ZZh/1N/JvYzfI4m362QmM6h3YmLiOORr6NKjX+K9EuzMWhq5qXm4YhXdwZdKOeV77LJDXLvqZU0bohiqI4fGBRSGqHCCHE9UuSIKJMuWfOoObno/PxQS8thgHQaSDAS0NWnoXiL62C/bUVbuH367FfmfxzQe2ahyMfpmvDrlcXaC1lMWWQ8OdDWI3p6AM6ENRztiRAhHAioN3jZJ1ZQU7cOowXj6L3a4MxKclxoKqSFxeHKT0dV//y1RcRdduItiP4cveXdsVaI4IiWDF+hV2R1ajFUVc8BmBPwh4GfTGI+zvdz+M9HsfXrXbVA3PRabi3n0eJdUPCA7Xc3EHPsr/yyHLygYXUDhFCiOuXvOsQZco+fhwo6Aoja2wLKIrCiO7uDgkQgJEVaOEXnxHPgaQDTF83HatqRa/VM7jFYOIz4isn4BpkzonHkHbQ9pWXuo/4P+7HnHkanUcoof3mo9FVvI2wuLbkG4388sQT/PLEE+Qbncx1v864+jbHq3FBa/e0Qx8DBQWqcfY7xmrl0LPPEvv115gvXqzGKEVNKFqs9Y1BbxARFGG3/WrGtKjXgu5h3TFZTCzYtYABCwawIHoBG09v5NYvbmVbzLZqucaytA9zoUlgwczLJoFaXr/Hh77t9LhoISbFwsINueQaHf9SK0CAp6bCH1gIIYSo26QmSCmkJkiBk7Nnk7lvH43uvZcGt91W0+HUGqpasCQmJuVyC78mFSzI1vy95iXuq+k12VejtGKQAKH9F+IZ2r96gxK1miknh5leXgBMy87G1dOzhiOqecb0I5z7dQigEP7/7N13eBRl18Dh32zNpickJAFC6D0oCtKLIoIoJSoqKPby2hABC75+oiiCKEXsoi/2AipNKSJIVSQUldBBAiSk97bJlvn+WLIQUkhCshuSc1/XXpCZZ2fOUnZnz5znPDeuI+9QuqMnyJleIMW/ejRpgvn0aQAUvR6fzp0pSk6mMCUFY2ioNE+th1zRiPX1za9zONVxM0Sv0WOxW+pM81Qou29IToGdjTGF/BZjLrMKBKR3iBBC1EeV/f4uOXBRIdVuJ+9MJYj0AylJURS6NNeX2DaqGlUgAE/2KXuZy7nD51YrtrrCVphRbgIEQOsR5MJohLg0GQM64tX0WkAlff8HBPToQcsJEzCFh6Po9ZjCw2n15JN0nDWLNs8+i1fbtqgWC9l//YX59GlUiwVzXBzHFywgIzra3S9H1CCD1uD8zFEUpVRy42LGKIrCoFaD+Omun5zjiqtGYpJjGP3laAYsdH+j9OK+IcUJEHD0+xjRw8Tr4/0ZP9CE/pw2XYriuGHROVxfxtGEEEI0BNITRFTIHB+PLT8fjdGIZ0SEu8OpU1RVZd/JIufP1b2oSs1L5du/vy21PTIkkpEdR15UjEKI+iGw82Pkxf9KzvGlNOriWAWmrKoO3y5d8OncmX2TJ1OUknJ2hzRPFdWk1WiZO3wuk1ZNKrWvrifq9TqFAZ1NBHhrWfCzo3eIqsLwK6t3w0IIIUT9IJUgokLF/UC82rRB0VZ/xZP66EiCldgUOxoNNPbTENXLs8oXVRabhSdWPkFSXhJGrZHOjTuXO29bCNFweQRdjmdof1BtpO//oMKxiqKU3RNEVTEnJNROgKJeG9lxpPNz6Vz5lnw3RFN1XZrrCW909hrm3BsYQgghGh5JgogK5cpUmHL98pcZgH4djMy4o2QpbmW9vvl1dsTtwNvgzY93/Mjy8csZd9k4lt65lMXjFpdZtiyEaJgCuzwOQM6/32PJrziZUV7zVI1ej2q3l/EMIcp3fvPUYK9gAP5v3f+x9shad4ZWKYqicEsfT+eSuJv2F/HXcUmECCFEQyVJEFGhvEOHAMfKMOKsxAwbf8c6LgqHXOZRrWOsOLCCRbsWATD7+tl0aNzhgvO2LzXSd1mImmNqfBWmxleh2ovIPLCwwrFhUVFnm6aew5afz8lFi+T/pqgSg9bA4nGLWXrnUsZdNo7fH/6dmzvfjIrKxJ8msv3kdneHeEGdwvW8fpc/113u+Mz+7Lc8MvMkISiEEA2RJEFEmTKio9n37LMUpaUBYMnOdnNEdcsvfzuqQC5roSc0oOrThA4kH2Dq2qkAPNLzEYa2HVqj8dUVuSd/LnefojGiNQa4MBohLn0BnR3VIFlHv8FakFL+uPObpzZvTuPrrwdFIW3jRuK/+koSIaJKzm2eqtFoeG3oawxpM4QiWxEPL3uY/cn73Rxh5UT1NNE8SEuuWeV/63Oxy/8DIYRocKQxqiglIzrasfziOU68/z4avV4a6gHZ+Xb+OORY8WRot6pXgWQWZPLI8kcwW830b9Gfp/o+VdMh1gl5CZvJPPARAAGdH8M7vOTyylpjAHov9y6tKMSlxjO0Hwb/jhRlHiD1r9fxb3+Pc9/5/6fKap5qataMEwsXkrx2LRqTiSY33+yq0EU9o9PoeOvGt7j3+3v5M+5P7v3+XhaPW0yEf91uoq7TKjwwxJtXl2RxIM7Kur/MDO1mcndYQgghXEgqQUQpCUuXlp5LfmZVAQG/xZix2qBliJY2oVXLI9rsNib+PJFTWacI9wtn/g3z0WrqX8NZS+4pErdNAFR8W99O0GVT8AjsUuIhCRAhqs6af5qirCMA5Bz/gVNrRjgfJ1YOxpIXX+HzGw0YQLO77gIgcdkykn4uv1pLiAsx6ox8GPUhHYM7kpqfyj1L7iElr/wKpboiLEDLbf08AVj6ZwEnUqxujkgIIYQrSRJElFKYmOiYS34uWVUAgEKLym97z1SBXF61Jfa2ndhGz/d6siV2Cx46D94f9T7+Jv9aitR97NYCErb8B3tRFsZGlxHc/SV3hyQuIYqiENypE8GdOskSlmWwFWaAWvYXNtVe6NgPWPLiMafHlHpY8uJpPGQITW69FYD4b7/lxCefsP/559lz333sf/55MqKjXfZ6xKXPx+jDolsW0dyvOSezTnLv9/ey7sg6hv5vKNtObCv3edtObLvgmNrUv6ORbi312OywcF0uhRaZFiOEEA2FTIcRpRhDQzHHxZVMhCgKHmFh7guqjvj9YCF5hSrBvhq6tazcajDxWfGkF6Tzwi8vkGF2fEF5tNej+Bp9azNUt1BVleQdz1OYsR+tsRFh/d9HozW6OyxxCdF7evLovn3uDuOSpdqKsOTFc2LlYFR7Yan9isZIxIj1hI4Ygc1sJmnFCtI2bnTuN8fFOaZDTpgg0x9FpQV7BfPpmE+5+cubOZBygEmrJpFvyefV317lmf7P4OfhR6hPKACJOYlkmbOYvWU2R9OPMnvzbF4d8iqBpkCa+rmuQlBRFO662ovjyVkkZdr55NdcbuheclqMj0njXFFGCCFE/aGo0hmtXNnZ2fj5+ZGVlYWvb/37wlqeUj1BFAVUlVZPPol/9+7uC8zN7HaVF77OIiXbzrj+nlwdWbl+IK3fbF3uvmNTjtVUeG5hyYt33nkGyDmx8kwfEA1NB3+FZ0gv9wUnRD1kTo/h1JoR5Q9QdOh9WmDJPlrukPBhK/EI7IKqqvzz6KPYcnPPO4aCKTycjjNm1FDUoqGo6POuMtzxmRhzsoi3fsotc5+vp8Ks8f7otVKVJoQQl4LKfn+X9LYoJaBHD8Lvu8/5s0ezZg0+AQKw57iFlGw73h4KfTpUvrph7vC5Vdp+qSi+23xuT4LiRqgoCnrvcPcGKERDpForTICcS1EU7IWlq0Vk+qOorvI+13QaHUadEaPOiE5TdhGyuz4TO4fr8TOVTnIoQKCXBp1cKQshRL0jb+2iTKamjpJUQ1AQnV57rcEnQFRVZe2eAgAGdTFi1Ff+rlDLgJaltkWGRDKy48gai88dbIUZZZbbA6DaSlSICCFcI2zAxzS67JlKjzeGhpZuhA2odjupGzZgt0rDSFF5IzuOJDIkssS2yJBIDj51kP0T97N/4n4OPnWw1BgvvZfbloovnhZzPhUY1bNqvb+EEEJcGiQJIspUlJYGgKFRIzdHUjccTbRyPNmGTgtXd6nasrhztznubvl7+PPKkFecF38Wu6XG4xRC1G9aYwCKpuxKNEVjxBjQAc+w/pU+XlhUlKP/0/lf9Gw2Ti5axP6nnyZ140Yytm+X5qnigoo/1yJDIsv9vDt3zCM9H0GjaMiz5PHUz09hs9tcHzQQGaGnaWDJldqaB2npHF653l9CCCEuLdIYVZTJmQQJCnJzJHXD2j1mAPq0N+LrWfnc4c74nWyJ3YJW0fLDHT/QIqAFY7uOxWK3YNAaaivcGnF+v49iWmMAqtVMxv4P3BCVEA2b3qspESPWl/t/U+/V9IJVWJbsf/EI7AI4pj8yYQKJy5ZhTkjAIyyMkBEjsGZlkbhyJUWpqZz85JMSz5fmqaI8Bq2BxeMWo9foURSlzM+788f0bt6b+3+4n1+O/sK09dN45dpXXF59oSgKt/QxlegNYrGpmC1gqtsf1UIIIapBkiCiTJYzSRC9VIKQmGHj71gLCjDksspXgaiqytytjiqQMZFjaBHQAnBcbF0KCZDyVpdwzJSWfspCuIveqyl6r/JX0SiuFilvulryzmkYAztj8HU0sQzo0aPMZEbQ1VeTsn498d9+C3b72R1nKkcSly2TJIgo5dzPt/I+787d1jeiL/NvnM/jKx7nm7+/oZGpEU/1e8olsZ6rc7ieFsFaYlNsKAokZNiZtyKbiSN88DRK4bQQQtQn8q4uytTQp8Ok59g4kWLlRIqVH7fnA9C2iQ5DFXqB/H7yd/489ScGrYHHej1WW6HWigr7faACCh6NZeUXIeqi4mqR8GErSzyaDv4Wg19b7EWZxG8YjyU3rsLjaAwGQq6/HkVTxqWCqmKOj6+lVyAammHthjF9yHQA3tn+Dp/v/tzlMSiKQlQvT8ICNIzt74mXUeF4so05K3LIM9svfAAhhBCXDEmCiDI15CSIxaYy4/tsXl3ieOw57pi/fPi0lRnfZ2GxXbgKQlVV5m2dB8DYy8bSxLdJrcbsamEDFxLa+80KexNojQEujkoIUUzv1RSPwC4lHp4hPWk6+BsMvm2w5icQv+EOrPlJFzyWMSys7OapNhv/vvWWrCQjasS4y8Yxse9EAKZvmM7KgytdHkOncD3Tx/pzdRcPpozywcekcDLFxpvLc8gpkESIEELUFzIdRpSpISdBdBoI9NaQU2ArMemjKsvlbfx3I3sS9uCh8+CRno/UVqhuozOFVKo3gRCibtF5NKLpNV9wat2tWHJPErf+dhp3fxWN0a/EuHP/D4dFRTl6gCjK2SaqquPdMXPnTjL37CH46qsxtWhB8tq1FCYmYgwNJSwqSqbLiCp5vNfjpOWn8cWeL3h61dPEZ8WzdN9SXhz8In0j+ro0lmZBOqaM8mXO8mzi0mzM+jGbcQM88fYoeRHgY9IQ6O3Ylp5jI8dc+kbJuWOEEEK4n6KqqkzuL0d2djZ+fn5kZWXh6+vr7nBcxmY28/eDDwJw2UcfoTWZ3ByR68WcLCrRIK3Ykzd606V5xf087Kqd0V+MZl/yPh7q8RDPDny2tsKsFaqqkhz9AtlHvy53TPiwlc7GikLUJEtBAV9dfz0Ad6xejb4Bvv+4giX3FKfW3oytMKXM/YrGSMSI9c5ESEZ0dInmqWFRURhDQ4n/7juy//qrjAM4EiUtpXmqqCK7aufhZQ+z4dgGFBRUVLqEdOHVIa8SaAqkqZ9rE+yJGTbeXJ5FVn7Z+309FWaN9wfguc8zyS4ofVldPEavleV2hRCiNlX2+7tUgohSiqtAtJ6eDTIBAmDUlWz/qSgQUcnl8n458gv7kvfhbfDmoaseqtU4a5pqtzoSIMe+c3coooFS7XZObNrk/L2oHXrvcIJ7TCdxa9mVaqq9EFthhjMJUl7z1DaTJ5Ozfz9H33wT1XLOst/SPFVUk0bRsOHYBgDUM5/CMUkxjP5yNADHphxzaTyhAVqeHu3LtG+zsZ33lnR+hejFVpEKIYRwDXlLFqU09JVhjiVaWPBzbomLGFWFUT1NF1y2z2a3MX/bfADuvfJeAkyXTl8MuyWP05sfOpMAUUApO0cq/T6EqB/03s1q5Dg+nTqVvUNVKYiLoyij4iV7hTjf3OFzy9z+VF/XrxoDEOKv465BnqW2q0Bqjp3nvsjiuS+ySM2xl1o7TaVy1w9CCCFcRypBRCkNuR/IsUQr81fmYLZAuyZazEVwMtVGi+DKVYH8dPAnjqQdwc/Dj/uuvM8FEVePJS++RC8PW2EGKbumY8k+iqL1ILTvAowBnaTfhxANmGq3VnqsMTQUc1ycs1eIk93OvkmTCOzbl5AbbqAgLo6EpUsv2DckIzq6UuNE/TSy40gW7VrE3qS9JbbP2zaP0zmnmdxvMo08XXuN0ru9kXV/FxKXZiuxPdesUtGy8SH+mkpdPwghhHAdSYKIUhpqEuTfJCtv/eRIgLRvouOJG3w4lmjl2615RPXyvOBdHIvNwlu/vwXAgz0exNejbvaRseTFc2Ll4HKXwA3t8xbezYYASLJDiAYsccsjNLpsCh7BPbBbskvtr0zzVGNYGIUJCaRt2kTamWlOxcxxcY7nnNc3JCM6usSxyhsn6i+L3TG1KjIkklu73srXf31NQk4CmeZMvvvnO1YfWs1TfZ9i3OXj0GlccymrKAo39zaV6Bc2tp8nrUNLnv9YopVvtp5tIJKWbefwaSvtm0oiRAgh6oqL+uSw2Wx89dVX/PLLLyQlJTF79my6detGRkYGK1euZPDgwTRtKl+iLjUNMQlyPMlRAVJQpNLuTALEqFecy+VVxtL9SzmReYJGno24q9tdtRvwRbAVZpSbAAHQedWv5XyFENVjLUgkafuUcvef2zw1oEcPmDChVPNU/+7dyT18mKSffyZr9+6SBzhTNXLio49I37LFuTnnwIES+6W/SMNj0BpYPG4xeo0eRVEY23UsFruFfxL+4eUNL7M/eT8vb3iZb//5lps638SSvUtcsoJM53A9LYK1xKY4KkSvjjSWukHSPFjLH4cKiU2xYTJAQREs+DmHJ2/0oV0TSYQIIURdUO0kSH5+Ptdddx2///47Xl5e5Ofnk3Fm3q+vry/PPfcc9913H6+++mqNBStco74nQc5fwu50uo2vNuVRaIW2YWcTIFVRaC3k7d/fBuCRno/gZfCq0ZiFEKKmaY0BKBpjmUlRRWPAr8P9ZB35EtWSU+bzK9s81btdO7zbtWPPvfeiWktPsbGbzWTt2VNxsKqKOSGhEq9K1BcG7dmV2BRFwaA10L1Zd5bduYxv//mWN7e8yaHUQ8zcNBOAGb/N4PVhr9fqCjKKohDVy7PCCtFzx4zp7cmve83sP2VlwU85TBzhQ5swSYQIIYS7VTsJ8tJLL7Fz506WLl1Knz59CAkJce7TarXcdNNNrF27VpIgl6D6nASx2FRmfJ9d5hJ2Wg08cr03HlVMgGw7sY1JP08iNT+VUO9Qxl02rqbCFUKIWqP3akrEiPUV9v7xbjKIuF9vq5HzGcPCyuwbog8MJCwqyvlzwo8/YimrmaqqkrZlC4F9+qBotTUSk7j0aDVa7rj8Dl789cUS2w+lHnLJCjKVqRA9d0z7ZnreWZXDgTjHlNunRvrSKkRmowshhDtVe3WYJUuW8PDDDzNq1Cg0mtKHadOmDbGxsRcTm3AD1W53rg5TH5MgOo1jCbuy0hxNA7V4GyufAInPimdv4l5mbZpFan4qAKM7jSY1L7WGohVCiNql92qKR2CXUo/i6g5FV3pFjHNZck5U+lxhUVFn+4WA89fw8eMJGjTI+Wg2fnyJ/cVUq5UTH33E/ueeI23bNjJ27GD/88+z57772P/882RER1c6FnHpK28FmUaejfjt2G8ujqZ8Bp3CY9f70L6JDrMF5q/MITa58k2HhRBC1LxqJ0FOnz5N165dy93v6elJTk7ZJbSi7rJmZ6PabKAo6APq3zKoiqIwqqepzD7uUb2qtoTdgIUDGP3laPYn73du+2DHBwxYOKAGIhVCiLovcdsTJP35HObUvzGnx5R6WPLinWMDevSg5YQJmMLDUfR6TOHhtHrySfy7dy9xzFLjmjenxSOP0OS229B6e1OYmMiJDz7g+NtvYz51CtVicTZPlURIwzGy40giQyJLbNNpdKTlp/HA0gd44McHiM2IdU9w5zHqFR6/wYe2YToKilTmrMhm++FCTqRYSzzSc+3uDlUIIRqEatfjNWrUiPj4+HL379u3jyZNpMHipaYo1VHFoA8MrLflxp2a6ZzNysBxszEiqHJL4J5r7vC5TFo1qcztdZmiNZW/T2NEa6x/yS9x6dBotXS65Rbn70Vdp5J97Duyj31X5t5zm6dC+X1DzlfeuODBg0lZt47TP/wA9nO+MErz1Abn/BVkFv+zGLtqp2d4T77Y8wW//fsb205s4/7u9/Noz0fZk7CH6eunu6SBalk89AoTbvBh3sps/k2y8cmveaXG+HoqzBrvj15btWm5QgghqqbaSZDBgwezaNEipkwp3Tn+2LFj/O9//2N8cUmruGTU534gxfYctzoTIOC4dh7Vs2pVIOC4C/Xe9vc4mn7UuS0yJJKRHUfWVKi1wpyyAwCdKZTQ/h+gaM5+0Tx3yUsh3EHn4cGYJUvcHYY4o+LmqUZC+swjPWYBRZkHy3z+uc1TLXnxFfYfKXahcaEjR5KwdCmq/by75qqK+fTpKr5CcakqbwUZg9bA2MvGMn3DdLbEbuH9P99n8T+LMRlMxGXFMXvzbF4d8mqtNlAtj4dB4ckbfXj6s0yKzpsRowCBXhp01a7RFkIIUVnVToJMmzaN7t270717d2677TYUReHnn39m9erVfPTRRxiNRqZOnVqTsQoXqO9JkPxCO99scdx98fNUyMpXaRFc9SoQcNyFyjA7LtS7hnZFPdPsr/girC5SVZXMw58D4N/hPkxBl7k5IiFEXVaZ5qk6r2bErS0/+Zux/yM8grqR+tcssBeV2n9utYglL54TKweXm3QpHldek1XVauX4e+/R5JZbMDZuXI1XLC4lZa0gA9AqsBWLbl7Er8d+5T/L/kNaQRoUOMbFJMW4pIFqeTyNGu4b7MUHa0tWgqhU74aMEEKIqqt2EqRNmzasX7+e++67j5dffhmAefPmARAZGcnnn39OeHh4zUQpXKa+J0GW/llAVr5KYz8Nt/f1ZMkf+eUuc3chGkXjTHxM6DOBQS0H1ekECDiqQIoyD6JoPfBtfau7wxFCXAL0Xk0rrBC70Ptn7smV5J5cWe5+1V5I+t4F6EyNsRYkl5kAKR5XXFUSFhXF8QULHPMZi5utnnk/zvjjDzKjowkeMgRTeDhJq1dTmJiIMTSUsKgomS7TQCiKwpA2Q3h96Os8u/bZUvvdOXX1ilYGwoPMnEq1Obc1a1S9GzJCCCGq7qLW6Lryyiv5+++/iYmJ4cCBA6iqSrt27bj88strKDzhavU5CXIs0cKmGMfF9Z0DvejYTE9ki+onLH4/8TvpBekEmgLpF9GvxF2ouqq4CsSnxWi0Bj83RyOEaAi8I0ZQlHWk3CkzANn/Lq7SMQN69IAJE0hctgxzQgIeYWGERUVhCAoi/ttvydm3j+TVq0s8p7h5KhMmSCKkAbm5y818+deX7E3a69zm7qmriqJwUy8Tb/2U69yWZ7aTa1bxMUkliBBC1LZqJUHy8vLo2rUrTzzxBBMnTqRLly506dKlpmMTblBfkyBWm8oXG/NRgd7tDXRsdvF3W1YcWAHA8PbD0Wvr/t0bS34CuafWAuDf7i43RyOEaCgCOj4EwKk1I8od4938RrQejbCZ08g9+VO54+xFZ1edK695aptnnyVn716OzZ+ParGc3SHNUxuk4gaq7YPacyTtCHbVTm5RrtsrNzuH62kRrCU2xYZWAxl5KvN/ymHySB88jdIYRAghalO13mW9vLxIS0vD29u7puMRbmY5kwTRBwa6OZKate5vM/HpNrw9FMb08bzo4xVYCvjlyC8Adb4RarHsI1+DasPU+CqMAR3dHY4Qop4obp5alsquOBXQ6WEad3+JgE4PVzju9OYHyTjwEYXZx8tdkldRFHy7di37AKqKOSHhgvGI+qO4gerPd//MnZffCYCHzgOd5qKKoS+aoihE9fIkLEDDnQO98DEpnEyx8c6qXAot6oUPIIQQotqq/QnQq1cvdu3axQMPPFCT8Qg3shcVYc1x3GUzBAW5OZqak5Jl46edjo5oY/p44mO6+Dss64+tJ8+SRzPfZlzR5IqLPl5ts9sKyTr2DQB+7e52czRCiPqkMs1Ty9pXHao1j9Q9M2HPzDL3l2ieGhpaZvNUaZja8BRXfDzR+wl+3PcjB1IOsHz/cqI6R7k1rk7heqaP9QegeZCWN5fncCTBygdrc3nsem90slSuEELUimp/G5w1axZLlizh888/r8l4hBsVT4XReHig9bz4aom6QFVVvtqcT5EV2jfV0bt9zZS+rjzgaPI3ouOIS6KTe+7JVdjMaehMoXg3G+LucIQQ9YzeqykegV1KPYobqla2WuRC44K6PY/WWP50zeLmqQBhUVFnm6aew15YiDUvr6yni3ou0DOQR3o+AsCcrXMwW8xujuis5sE6nrjBG4MOYk5aeH9NLrFJFk6kWEs80nPtFz6YEEKIClW7EmTSpEn4+flx7733MmXKFFq1aoXneV+cFUVh/fr1Fx2kcI1z+4FcCl/sK2PH0SL2nbKg0ziaodbE68osyGTT8U0AjOo46qKP5wqZhz8DwK/tHSiaut+/RAhRv1SmWqSy4zyCriBu3S0XPOf5zVMNQUFYsrIoSk3l6Btv0PbZZ9GaTBf/4sQl5Z4r7uHLv74kISeBT3d/yn96/sfdITm1DdPzyDBv3v45l39OWPjnhKXUGF9PhVnj/dFLlYgQQlRbtZMg//77L4qi0Lx5cwCSkpJqLCjhHvWtKWqe2c53W/MBGH6liVB/bY0cd9XhVVjsFjo17kTboLY1cszaZE79i8K0v1E0Bnzb3O7ucIQQDdSFltqt7DhFW3alSDFz2l48ArtgyYvH1NpEy8ljnfsKkxI58eGX5B87xrF582gzZQoaQ91e1UvULA+9B5P7TWbK6im8/+f7jIkcQyPPunPd06W5gQeu9eKjdaWrlRQg0EuDTvqmCiHERal2EiQ2NrYGwxB1QX1IgqTn2MgxO+Z//7yzgJwClUbeCj3b1dxFrnMqTIfyVzqoS4qXxfVufgM6j/rT60XUT0V5ebzVogUAT8bGYvDycm9A4pKTEv082ce/pzB9L9hL30k39TSQv+0qcg8c4N8FC2g1cSIanXubZArXGtVpFIt2LWJf8j7e+eMdpg2e5u6QSujR1si/SVZ+/aewxHYVGNXTVG+qdYUQwl0klyycnCvDXKJJEItNZcb32by6xPHYc9xx8ZuWq/L60mwstovvtn46+zQ74nagoDCiY91PgljNqeSe/BmQZXHFpSM/NZX81FR3hyEuVYqOwtTdZSZAAFCLCL/vNhSDgey//yb2/fdRbTbXxijcSqNoeG7QcwB8/ffXHE8/7uaISru1rycB3meTHYoCLYK1dA6XKa1CCHGxLjoJUlBQwKpVq3jnnXd45513WL16NQUFBTURm3CxS70SRKeBQG8N598fqcny0eIqkKvCryLMJ+ziD1jLso9+i2ovwtjoMjyCLnd3OEIIcdEu1Dy12bXf4tVsWIXH8GzRgtZPPomi1ZK5Ywf/PPooe+67j/3PP09GdHRthC3qmD7N+zCo5SCsditvbHnD3eGUoigKdw06WwmnqlIFIoQQNeWi6j+/+eYbnnjiCTIyMlDPLEGnKAoBAQG8/fbbjB079gJHEHXJpZ4EURSFUVeZeOvn3BLba7J8dMXBFcCl0RBVtVvJOvIVIFUgQoj6ozLNUxWtkby4NRUex7drV4KHDiV51Sps+Y7+Uea4OI4vWAATJjgaq4p67dmBz7I5djNrj6xlZ/xOujft7u6QSugcrifYV0NKth2NAm3DZNqWEELUhGrfG1+3bh133nknBoOBl19+maVLl/Ljjz/y8ssvYzAYGD9+PL/++mtNxipqkaqql3wSBOBYYsny55osHz2UcoiDKQcxaA0Ma1fxXUZ3seTFY06PwZweQ8bBj7EWJKLR+2Fs1M3doQkhRI250JK8lZW9d2/JDWdu6CQsXVpToYo6rF1QO8ZEjgFg1sZZzht6dYWiKNwxwBONBuwq7DhSzhQvIYQQVVLtlPLMmTNp2bIl0dHRBAQEOLePHj2aRx99lKuuuoqZM2dy7bXX1kigonZZs7NRLRZQFPSBge4Op1o2xpj5add5TcRqsHx0xQFHFcjAlgPx8/C76OPVNEtePCdWDka1l/wzsFuyOLXqeiJGrK/yFwQhhKiPVLsVgMLExDL3m+PiyI6JwadzZ5l+UM9N7DORlQdWsidhD6sPr2Z4++HuDqmEzs0N3NTTxPd/FLD+HzP9Ohrk36QQQlykaleC7Ny5k/vvv79EAqRYYGAg9913H9Eyr/aSUVwFovf3vyS75O86VsTXmx3lzDdcaaRFsGM53JqqArGrdmcSpK5OhbEVZpRKgBRT7YVllo4LIUR9VFHfEICMfe+h2m0YQ0MdJYPnU1WOvv46h199lZz9+2sxUuFujb0b82CPBwF4ZcMrXPe/69h2YpuboyqpX0cjBh3Ep9s4fNrq7nCEEOKSV+1vuzabDQ8Pj3L3m0wmbNJt/ZJxKU+FORBn4eN1uajAgE5GRl3lSbsmVr7dmkdUL88auWOyO343p3NO423w5upWV1980EIIIWpNeX1DCpJ3kLr7NfLi15Ec/QKho6OIffttRyJEVZ2/+l5+OTn79pF3+DBHZs7Eu2NHfDp1ImPHDgoTEzGGhhIWFVWqb0hGdDQJS5e6ZIyoOde3u57Pdn9Gcl4yyXnJzN48m1eHvEqgKZCmfu6voPTy0NC7vZFN+wpZ/4+Z9k1lhRghhLgYilrNCZA9evRAURS2bduGXl/yzdhisdC3b19UVb2kq0Gys7Px8/MjKysLX19fd4dTq5LXrCHuq6/w79mTVo8/7u5wKu1kipU3lmVjtsAVrfQ8fJ03Gk3Nl4n+37r/4+u/v+bmzjcz+/rZNX78mpBzag2JWx4pd3/4sJV4BHZxYURCVF1RXh4zvb0BmJqbi8HL6wLPEKJqck78ROLvT4Jqx7/D/Wit15G0fDnmhAQ8wsIIi4rCv3t3itLTSfrpJ1J/+w3VWvbdd78rr8QjzLFSmDkhgaxdu2ptTEtp1lprWr/Zutx9x6Ycc2Ek5TudbmPat1koCrx2hx9Bvlp3hySEEHVOZb+/V7sS5NFHH+X+++9n0KBBPPPMM3Tq1AmAffv28eabb7Jr1y4++eST6h5euNilWAmSnGVj/k85mC3QvomOB66tnQRIka2IVYdWATCqk/umwljy4sue0qKqZP+7mKwjX7s+KCGEuMT4RNyI3VpA8p/PkHnwE3xb59Ny8rgSYyx58RgCmxJ+112E3HADB154AVtubqljZe3aRdYFzldTYxJ+/FGSILVk7vC5TFo1qcztdUWTQC2dmunYH2flt5hCxvTxdHdIQghxyap2EuTee+/lyJEjvP7669x0002l9j/33HPcc889FxObcCFnEqSONkVNz7GRYz5btJRrtvPphjxyClTCg7Q8NtwHva52GoVtjd1KpjmTYK9geoX3qpVzXEh5TU+FqG+0ej0Dp01z/l6I2uDXegzW/NOk751P9rFvyD72TYn9isbobCZtaNQIe2E5770aDY2HDgUgee1asNtrbYw5Lo7kNWsIGjwYjfzfqFEjO45k0a5F7E06u1qQn4cfIzqMcGNUpV3T1YP9cblsPVDIyB4mjHppkCqEENVxUR0wX3vtNe69916WL1/Ov//+i6qqtG7dmtGjR9OmTZuailG4QF2uBLHYVGZ8n012QemZW4oCj13vjclQexcCyw8sB+DGDjei1bin/LSipqcAet82BEY+SfIfU8ocp2iMaI2lmxgLUddoDQYGvfSSu8MQDYBX08Gk751f5r7iZtLFK2oZQ0Mxx8U5l9AFQFEwNWtGs3GOKpLsmJjaG3NG3FdfkfzLLzQZM4aAnj1RNNXuby/OYbE7lp6NDImkZ3hPPt75MVnmLD7e+TEPXfWQm6M7KzJCT7CvhpRsO38cKmRQl/J78wkhhCjfRS8D0rZtW6ZMmVITsQg3ciZBgoLcHElpOg0EemvIKbBx/iVhkwAtgd61dxH469Ff+fngz4DjTlFdFdJ7LqZGkZiCupU5ZUZrDJDlcYUQoprCoqI4vmBBqQaqYVFRLhsTNHgwWbt2UZSSQux775G8ejV+3bqRER0tjVgvkkFrYPG4xeg1ehRFoYlPE6b/Np03t7xJZGgkvZv3dneIAGgUhWsiPfhuWz4b9poZ2Nkoy+UKIUQ1VLsx6vHjx4mJiWHEiLJLBVeuXElkZCQtWrS4mPjcqqE0RrVbLPx1330AdH3vPXQ+Pm6OqLSYk0W89VPp+dhP3uhNl+aGGj9ffFY86QXpPL7iceKy4xwXSGMXu61TvDk9hlNryi/LlaanQghRNRd6Xw0btAjvJoOcP2dER5O4bFmpBqrnqu0xNrOZ5DVrSPr5Z+xmc5lxB/Tti6lZMwAK4uLI2FZ6udcKx5xJujTkRqyqqjJl9RSW7V9GoCmQ5eOX08S3ibvDAiC/0M4zn2dSaIGnRvjQKVymRgkhRLHKfn+vdhJk3LhxnDp1ii1btpS5f9CgQYSHh/PFF19U5/B1QkNJghQmJbFvyhQUg4HLP/64Tt5VOJViYcYPOdjOTJNWgIhgLc/f4lsr8da1TvGSBBFCiJp1ofdVFB1+bccR2OlRVNV6wSq78ppX18qYrCz2P/dcmc1aa4pHeDidXnut1o5f15ktZsZ8M4b9yfuJDInku7HfYdQZ3R0WAN9syWPD3kIiI/RMuKHu3bgSQgh3qfXVYbZu3cpDD5U/T/K6667jo48+qu7hhQud2w+kLiZAdh4tYtGGXGcCBEAFRvU01Vq8da1TvGorcst5hRCiwVKtZB3+nKwj3wB2UG2lhhQ3UAXKbV5d02P0Xk3R+/mV36xVUWjUvz8AaVu2lNlbpDJjzHFx5B46hHf79mWfp57z0Hvw/qj3GfXFKPYm7WXar9OYOXRmnbhOujrSgw17C9l7wkJSpo0Qf1kuVwghqqLazRSSk5MJDQ0td3/jxo1JSkqq7uGFC9XVpqh2u8qP2/P58JdciqzQsamW8CDHB32LYC2da7EEdFDLQeg0JXOEkSGRbusLkhP7Y7n7pOmpEEJUndYYgKIp+86+ojES0uctPIK7g2opMwECjgaqOceXkXN8WbnNq2tyzLkVIsbQUMfUlRKBK5jCw4l48EEiHnwQj2bNqjcGQFU5/OqrHJs3D/Pp02XGVN8182vGWze+hUbRsCRmCd/8882Fn+QCof5aIps7roF+iyl7WpQQQojyVbsSxN/fn2PHyp8WcPToUXzqYG8JUZq7kyDnL38LYC6yszLazKHTVgCuu9yDm3qZOBRv5duteUT18qzVuzHTN0zHardi0Bp4buBzLN23FHB0kDdoa74HSUUKUnaTddRx4RXU7b+YQkou0ytNT4UQour0Xk2JGLG+wqknPhEjyDzyJak7Xyz3OGn/vHnBc9XUmHPVdiNWn8hIcvbtI2v3brL++gufTp2wpKdTmJLSoJqn9mvRj8n9JvPGljeYvn46FpuFr//6mhcHv0jfiL5lPmfbiW1MXz+9wjEX65quRvaetLDtQCGjrvKs1VXyhBCivql2T5AxY8awceNG9u3bR+PGjUvsS0xMpEuXLgwYMIAffyz/DnZd11B6gpz45BPSNm4k7KabSlwYuYLFpvLc55llLn8LjpVh7rnGi57tXDcPd93Rdfxn2X9QUPj29m/p3qw7qqq6JQFitxZwcvVwLDmx+LSIIrSPe6bjCCFEQ3Wh3iHGwMsAKEz/u9bHBPeYgX/bcc6fa7sRq/n0aeK/+46s3btLBtLAmqeqqsr9P9zPpthN6DQ6rHYrXUK68OqQV0s0TC9uqv7CuheISYopc0xNxvTiN1kkZtq5vZ8ng7vKcrlCCFHrjVH/+usvevXqRUBAAJMmTaJr167O7fPmzSMzM5OtW7fS/bwP2sr45ptvePPNN9m/fz9eXl4MGTKEWbNmERERUeVj3XrrrSxZsoT27dtz8ODBKj23oSRBjsyeTc7evUQ8+CCNBgxw6blVVeW177M5kVJ6+VutBp67yYcWjV3X+Tw9P51hnw4jLT+Nh3o8xLMDn3XZucuSsms6mYcWoTOF0vyGNWgNfm6NRwghGprKNKYGXDIGwLPJ1QRd/iwavbfLGrHGTJ5MUXJyyQFnptV0nDGjwnjri4oapt/S5RYAvo/5vtwxtdFU/aed+SzfYSbAW+HRYd7OClkfk4ZA72rPeBdCiEtWrTdGvfzyy/n++++59957efbZZ51vvKqqEhQUxJIlS6qVAHnnnXd44okn6Nu3L/PmzSM1NZX58+ezefNmoqOjadKk8kuU/fzzz/zwww+YTKYqx9GQuHM6jKIojOppKnP52/uv9XJpAgRg2vpppOWn0bZRWyb2nejSc58vP+kPMg8tAqBxz1mSABENgqqq5KemAuAZFFQnmhAKUXdoyT/9GydPbzxTjWEvNaI2GrFaMkonSVBVzAkJF/dyLiHlNUyHipMfxc+taRabyoa9jr+3jFyVGd/nOPf5eirMGu+PXivvn0IIUZZqJ0EAbrzxRk6ePMnatWs5cuQIqqrSvn17rrvuumolHtLS0pg6dSpXXHEFGzduRKdzhDds2DCuuuoqXnzxRT7++ONKHSs3N5dHH32URx99lJUrV1Y5loZCVVUsZ75w6IOC3BKDUQd6LVjO6TsXEayle2vXTj356eBPrDq0Cq2i5Y3r33DrUnh2Sy5J258BwLfNWLyaDHRbLEK4kiU/nzfPTLGcmpuLwcvLzRGJhq64gWp5iYLixtSuGNPk6k/JOvw5uadWl73qC44Gqrlxvzp/fzFjbIUZ6L2aYgwNxRwXV+qcik6H3WpFo7uoy8lLwsiOI1m0axF7k/Y6t4V6hzL+8vElbgR+8dcXJOYmOscEmAIY0aHi6p7q0GmgkbeGnIKSTXsVINBLg04KQYQQolwX/allMpkYPXo0mZmZrFq1isOHD6PX6xk+fHiVj7V8+XJyc3OZMGGCMwEC0L17dwYMGMDixYt57733MBgu/OX4hRdewGKxMGPGDEmCVMCWm4u9yLH8qiHAtSuMFBSpLN2ez8aYwlJTYUbX4vK3ZUnJS2Har9MAeKzXY0SGRrrs3GXGs+c1rHlx6LyaEdztebfGIoQQDVllGqgCLhvjGdKLrGPfk/zn0+XGnLrrpQu+rsqMKVaqeeoZ9oICjr/9Ni2feKLeJ0IsdgvgWCnu1q63svifxQDc1+M+Z7+wIlsRa46sIdgrmCubXsmnuz8loyCDr/7+ivHdxtdoPOVV0qrAKBdfQwkhxKWmSp9YS5cuZdGiRXzwwQclpqXs3r2bESNGkJiYiKqqKIrCNddcw6pVq9DrKz+dYceOHQD06dOn1L4+ffqwadMmDh486Ow/Up7o6Gjefvttvv766yr18igsLKSw8Owdkezs7Eo/91JVPBVG5+eHphLJpaoqa+UXgFMpVpZHF5CZ59jXp72eU2l2TqXaan352/Opqsrza58n05xJ58adebTXoy47d7Fz52QXpOwi+8xqMI3OzPsWQgjhPnqvphdchcuVY4wBHSo+hk8rACw5/17UmMLMg3gEdsG7UxPCH7qR1I0bKUpNxRAUhHeHDqRt+JOs3bs5/u67tHrsMZR6nAgxaA0sHrcYvUaPoiiM7Tq2VMP088cEeQbx5tY3mfHbDLqEdKFbk241GlPncD0tgrXEppytBtFpHNUgQgghylelT6vFixdz+PDhUn057r33XhISEhg7diy9e/dmxYoVrF+/nvfee48nn3yy0sePj48HoFmzZqX2FW+Li4urMAlitVp58MEHufbaa7ntttsqfW6AmTNn8vLLL1fpOZe62uwHYrGpzPg+u9yVXwCCfTWMH+RFx2Z69p+yuGT52/P9sO8HNvy7AYPWwBvXv4Fe69o+JJa8+HLnZCf/MQVTUDdZAlcIIUSlhfZ9C6i4yWplxiRvf5rsf3/AnLoL7BZ0rUF3pj+o2bIGrwF68jZfSdbOnRx/7z1aPvpovU+EFFMUpcwV487d9p+e/yEmKYY1R9bw2IrHWD5+OcFewTUWT1nVIFY7zP8pl6vaGritrye+njIvRgghzleld8Zdu3Zxww03lNi2e/du9u7dy6hRo/jqq694/PHHWbt2LVdccQWLFy+uUjD5+fkAGI2lezF4eHiUGFOeOXPmcOjQId57770qnRtg6tSpZGVlOR+nTp2q8jEuNUXp6QAYAgNr/Ng6DQR6a8q9IzH0ciMv3e5Hx2aOpEOncD3Tx/rTyYVVIKezT/PKhlcAmNhnIu2D27vs3MVshRkXnJMthBBCuJ4Gc/J2ODMVpBTVQrO7bkHR6ciMjub4Bx+g2mxlj22AFEXh9etfp01gG5Jyk3hi5RNYbOX8WVZTcTUIQPMgLddEGlAU2HGkiP/7Jos1u/OJTbZwIsVa4pGeW7qprhBCNBRVStcnJyfTpk2bEtu2bNmCoijcfffdzm2KonDzzTcze/bsKgXj6ekJOKalnN9YtaCgoMSYshw7doyXX36Z559/ntaty1/KrDxGo7HMBEx9VnSmKWptVIJUtPLLuP6eXB3p3jXtt8Zu5T/L/0OBpYBuYd14oMcDbo1HCCGEqAxXNWttcvVnZBz4iPzTG8qNxbt9e1pNaMW/b71F5p9/cjgjA1t+PoVJSRhDQwmLiiKgR48Sz8mIjiZh6VIKExPLHVNfeBu8eX/0+0R9GUV0XDQzN83kxWterLHjK4pCVC9Pvt2ax829PekUrqd3eytfbMzjZKqNH7abAXOp58kKMkKIhqxKSRC7vXTWODo6GoDevXuX2B4aGkpeXl6Vgmna1FHyHxcXR9u2bUvsq2iqTLHJkycTEBDAbbfdRmxsrHO71WrFYrEQGxuLyWQiJCSkSnHVZ5Yz02H0tbQ8bqdmOryMCnmFjikxCtA8WMugLu5LNsVnxZNekM5za5+jwFKAgsKDPR4kMSeRpn4y7UQIIUTd5spmrRq9V4VJEAC/bt1o+cQT/LtgAXmHDzu3m+PiOL5gAfZHHsH/iisAyNy9mxPvv+9sslo8hgkT6m0ipFVgK94c/ib/WfYfPtv9GV1DuzK60+gaO35xJW2xFo11PH+LL7/+XcAPf5hLNZ+XFWSEEA1dlZIgERER/P333yW2bdmyhaZNm5ZKLGRlZRFYxSkWPXr04MMPP+T3338vlQT5/fff8fb2pkOH8puBxcbGcvr0adq3L3tKQ8uWLRk6dChr1qypUlz1mbMnSC0tj/v7IYszAQKOruWuXvnlfAMWDijxs4rKoysczVCPTTnm8ngsuXEuP6cQQohLmysbsVaG/5VXYggMdFaYAs6VZE68/z4nzn9C8SozqgqKQuKyZfU2CQIwpM0QHuv1GO9uf5f//vJf2gW1o1PjTrV2Pq1GYWg3R0+Q/60vOZVcVpARQjR0VcoBX3/99Xz11VesWrWK/Px85s+fz6lTpxg1alSpsbt27aJ58+ZVCmbUqFF4enqyYMECrFarc/vOnTvZvHkzt956q3N53ISEBA4ePFiiR8i8efNYunRpqUdwcDBNmzZl6dKlTJs2rUox1Xe12Rg1Ps3KN1sc1UD+Xo4PWlev/FKWucPnVml7bbKZ00nd1bCa8QohhKg/VNvZqRaWrKxqHkTFnJBQQxHVXU/2eZKBLQditpp5ZNkjrDm8hqH/G8q2E9tq7Zy92hmJCNaW6s+2N7YIs6X8xvVCCFGfVSkJ8vTTT+Pl5cWIESPw8fFh8uTJ+Pn58fTTJdeqN5vN/PTTTwwYMKCcI5UtKCiI1157jd27dzNo0CA+/PBDZsyYwbBhwwgJCWH69OnOsVOnTqVjx47OZXUBrr76akaPHl3q4enpibe3N6NHjy41bachU61WLJmZQM0nQQotKh/+kkuR1dG0655rvAgL0Lh85ZeyXNXsKpTzLgciQyIZ2XGkS+Ow2wo5veVhrAWJ5Y45d962EEII4WrF/UfKk/LX69htjp4ixtBQxzSXcykKHs2acfnHH3P5xx/j0axZ6TGAR2hojcZdF2k1WubdMI8mPk2Iy47j2TXPcjT9KLM3z2Zv4l7is+Jr/JyKojC6p6nUlJgNMUW89G0W+0/VbKNWIYS4FFRpOkxwcDA7d+5k9uzZHDlyhFatWjFlypRSFR/bt2+nb9++jBkzpsoBPfnkkwQFBTFnzhwmTpyIp6cnQ4YMYebMmc6eIaJmFGVkgKqi6PXofHxq9Nhfbc4jIcOOn6fCfYO98PXUMH1s6aXk3OGDHR+gouKp92TqwKks3utYxchit5S53F1tUFWV5D+nYk7ZiUbvQ2j/99Ea/EqNO3fethANgc5o5JYzK4vpGlijaiHqovL6jxRmHiQ5+kUKU3aSuPUxwvq/T1hUlKO/x5l+H8W/Nrn5ZjRn/j+H3XRTyTFnaH19UVXV7TdKapufhx+nc04DkFvkaBwfkxTD6C9HA7UzLbd4BZnYFBstgrWMusrEl5vzScuxM29lDn07GBhymQfWMhaM8TFpCPSW5iFCiPpFUVVVauHKkZ2djZ+fH1lZWfj6+ro7nBqXc/AgR2bMwBgSQuc336yx4247WMinG/JQFJg80of2Td07/eVcybnJDPp4EIXWQj675TP6teiHqqouTYAApMe8Q9o/c0DR0mTQ//AKq1rVlBBCCOFu+UnbOb3xHlRbId4RIwjtPY/MXbtJXLYMc0ICHmFhhEVF4d+9e4nnZURHO8fo/fycfURCRoyg6a23uuOluNTy/cuZtGpSqe1zh89lVKfSU8xrwv5TFr7dmsft/bzoFK7HbFFZuj2f3/YWouJollrWFwJZRUYIcSmp7Pf3KlWCiPqlNvqBnE638fVmRx+QkT1MdSoBArAweiGF1kKuaHIFfSP6Ao5SUVcmQHJO/ORIgADB3V+SBIgQQohLkmdIL8L6vc/pLQ+Te2IliaqKf8eHaDl5bIlxlrx4Z1WjJS8eU2tTiTGZO3eSsORnklauROfjQ8j117v0dbjayI4jWbRrEXuT9jq3dQzuWKvTcs9fQcZDrzC2vxfd2xj4bEMuSVmlUyCyiowQor6SJEgDVtPL4xZaVD5Y6+gD0rGZjuFXeNTIcWtKal4qX//9NQBP9H7CZSW3lrx4ZxlxUeZBknY8D4BvqzH4t73TJTEIIYQQtcGr6dWE9plP4tbHyT35E7knfyo1RtEYiRixHoATKwej2gtLjfG+Wkfub5cT//XX6Hx9adS3b63H7i4Wu6MPR5eQLqTkpZCUm8TJrJOYrWZMepNLY2kbpmfabf78b0MuO4+W7A8iq8gIIeorye02YDVdCfLNljwSMmz4eSo8cK03Gk3d+tD8KPojzFYzl4ddTv8W/V1yTktePCdWDubUmhGcWjOCpO1Pw5mLn5zYFVjyar4JmhBCCOFKPs2HExj5ZLn7VXshtsIMbIUZZSZAHKw0GnQVACcWLiTr779rIdK6waA1sHjcYpbduYwvxnyBh9aDvKI8Pt75sVvi0esUHhriTVjA2a8FilI3VvQTQojaIEmQBqx4Dm51kyDpOTZOpFg5kWJlZXQ+2w4WAXBbP8e69HVJal4qX//lqAJ5vPfjLrurUdEFX/FFoRBCCHGp82o6uML9CZseJGHTgxWOCR46lMC+fcFm49j8+cRMnsye++5j//PPkxEdXZPhup1Ba0BRFFo3as0r170CwILfF/DnqT/dEo+iKNza19P5s6pKFYgQov6qW99UhUtdTCWIxaYy4/tsXl3ieKyINjv3fbs1H4utbvXb/WTnJxRYC4gMiWRQy0HuDkcIIYRoUKwFiRUuCQ+gaDREPPAApogIsFopSk5GtVgwx8VxfMGCepcIKXZT55u4qfNN2FU7T/38FGn5aW6Jo3O4niaBWufPgV7yNUEIUT/Ju1sD5kyCBAVV+bk6DQR6azj//kBdbKKVnp/Ol399CcATfVzXC0QIIYQQDo17vkHjnm9ccJyi06HabCU3nlluN3HZstoJrg54afBLtA5sTVJuEk+vfhq7WsZ6tbVMURRu6+uJ8cwMmF/+Nlf8BCGEuETVoa+qwpVs+fnYzY4PN0NgYJWfrygKo3qaSi2nVhebaP1v1//It+TTOaQz17S6xqXnlhWohRBCCDAGdMAY0KHCMardkfwoTEoqY6dKQXw8qt31yQFX8DJ4sWDEAow6I5uOb+KTnZ+4JY5O4XomjXQsK7n9cBHpObYLPEMIIS49kgRpoIqrQLTe3miMxmodo1MzXYmKj7rYRCujIIPPd38OuHZFmGLmZPfM7RVCCCFcSWsMQNGUfT2haIxojQEVjgFIj3kb1W7FGBrquKg4n83G/qlTSf/jj3qZDOkQ3IH/u/r/AHhzy5vsOb3HLXG0CtHRvqkOmx3WSTWIEKIekiVyG6iaWBnmaKIN6znXIHWxidaiXYvIs+TRMbgj17a+1qXnttsKyTy0qNz9xReFQoiz7DYbJ7dsAaB5//5otNoLPEMIURfovZoSMWJ9mQ2/tcYA9F5NAcocU5D0J6l/zSL/9HqS/phC6OixxL79jiMRcmYqDKqKxmik8PRpYt97j8Tly/Ht2pXsmBgKExMxhoYSFhVFQI8eLnm9teX2rrfz+8nfWXVoFU/+9CQr71qJn4efy+O4vpuJQ/E5bN5fyA3dTXh7yH1TIUT9IUmQBupiV4YB2HrAseqJt4dCrlmtc1UgWeYsPtv9GeCeKpDMg//Dmn8ajTGIJv3fQ9GZSuw/96JQCOFgNZv57OqrAZiam4vBy8vNEQkhKkvv1fSCn2tljfEI7ILeJ5yELY+Rc2I5Pi0UWjzxOEnLV2BOSMAjLIywqCh8Oncmee1aklevxhwfjzn+7DLzxc1TmTDhkk6EKIrCa9e9RkxiDCezTvLg0gfJLMhk2uBp9I3o67I4OoXraB6k5WSqjQ3/mBl5leeFnySEEJcISes2UBdbCVJQpLLrmGNJ3Buu9CAsQENUL886VQXy6a5PyS3KpV1QO4a0HeLSc1vzk0jf9w4AwVc8j6lxDzwCu5R4SAJECCGEcPBudh1h/d4GRUtO7DIK8j+ixVO30vHNSbScPBaPVh7Y7emEjR5N57lz0fn6ljzAmR5cCUuXuiH6muVj9OG/V/8XnaJjV/wujqUfY/bm2exN3Et8VvyFD1ADFEXh+iscN2827C3EbJEeZ0KI+kMqQRqgjOhoUjZscPx+xw68O3as8l2TnUcLKbJCqL+GwV09uPYy04Wf5ELZ5mwW7XZMRXmi9xNoFNfm+1L/fgPVmo9Ho274tBjl0nMLIYQQlyLv8GEEX/kSKTv/j/zTG8g/vaHEfkVjJGLEevReTbEVFJR5DPOpU2Tu3o1ft2516sZMVT287OESP8ckxTD6y9EAHJtyzCUxXNFKT2M/DclZdrbuL+Tayzxccl4hhKhtUgnSwGRER3N8wQLs+fkAWLOyOL5gARnR0VU6zraDjiqQvh2NdfIi45UNr5BTmENTn6YMazfMpec2p+4h5/gPAARf+SKKixMwQgghxKXKI+jycvep9kJnP5Fym6cC/86bx6Fp08j66y9UVSUjOpr9zz/PnvvuY//zz1f5mscd5g6fW6XttUGjURjazZH4+OUvM1abVIMIIeoH+XbWwCQsXVr6okFRSFy2rPLHyLBxLNGKRoFe7aq3skxtic+K589Tf7LswDIAFI3CvqR9LisfVVU7KbumA+DT8uYKL+aEEEIIUT1hUVFnm6aC81e/K69EYzCQf/w4x+bMYd/TT3N8wQLMcXGoFouzd0hdT4SM7DiSyJDIEtsiQyIZ2XGkS+Po3d6In6dCRp6dPw8XufTcQghRWyQJ0sAUJiY65806qSrmhIRKH2PbQUdD1C7N9fh71a1/QgMWDmDcd+Owq45la+Ky4hj95WgGLBzgkvPnxC7DnPYXis6LoMufcck5hRBCiIYi5/hS7NYCvDs1IfyhGzG19UIbaMbU1ovwh0fQ/MExdJ47l8bDh6MYDBQlJTmeWHztcyZxUpWbP+5gsVsAaNeonXObTbU5t7uKXqsw5Mw0mDV7CrCffw0phBCXoLr1DVbUujLLRxUFj7CwSj3falP540wSpG/HulUFAnDvFfeWud0V5aN2Sx6pf70OQGCXx9GZGtf6OYUQQoiGJPPQ/zi+vB+xKwaReeo5dK3X4NlzJ7rWa8g8+SwnVg4GXS7Nxo6ly5w5oCnjUreKN3/cwaA1sHjcYlbds4pOjTsBcGvkrRi0BpfHMqCzB55GhcRMO38dd20SRgghaoMkQRoYZ/loMUUBVXVsr4R9pyxkF6j4mBS6RtSd5XDBMRXm+5jvS213Vflo+v73sBUko/eOwL992ckYIYQQQlSf1qMx9sJ0UK1l7j+3b4je3x+PpmWsxFaFmz/uZNAaUBSFGzvcCMDqw6vdEofJoDCoi+PG1+rdBahSDSKEuMRJEqSBCejRg6a33+782RQeTqsnn8S/e/dKPX/rAUcVSK92RnTautMQ1WKz8OTPT5JTlINJZ6JLSBdeGfKKcz5tbZSPWvLiMafHYE6PITfuVzL2fwRAQKf/oNHWvSoZIdxh0KBBtGjRwt1h1An33HNPnWwkDbUbm6Io3HPPPZUa++mnn6IoChs3bqyVWETdpzUGoGjK/gxVNEaaXfsdAZ0frfTxyrzJU4WbP3XBiA4jANhxageJOYluiWFwpAc6DcQm29i0r5ATKVbnIz3X7paYhBCiumSJ3AbIdOYLiUfTpnScMaPSz8vOt7P3hCOZ0LeD68sxKzJv2zz2nN6Dj9GHH8f9SMvAliiKwtiuY7HYLTVePmrJi+fEysGo9sJS+1J2voRnWH/0XmXcfRKiHjCbzSxcuJDvv/+emJgYsrOzCQgIoFu3btxyyy3cddddGI0NMxG4ceNGNm7cyMSJE/H393d3OKICOTk5TJs2jSVLlpCcnExERAT33XcfU6ZMQae78OXR9u3befPNN/nrr79ISkpCURQiIiIYM2YMTz75JH5+fiXGP/fcc2zevJmjR4+SlZVF48aNueyyy3j66acZOHBgbb3MS47eqykRI9Y7qznOpTUGoPdqinf49WTse6/cY6i2szc+Anr0gAkTSFy2jIJTp0BVMYSE4HfFFbUSf21o4tuE7k27szN+Jz8f+pn7u9/v8hhMRgVFA9jhq835Jfb5eirMGu+Pvg7dHBNCiIpIEqQBsuXmAqDz9q7S87YfLsRmh5aNtTRtVHf+6Ww6vokPd3wIwKyhs2jVqJVzn6IotTJ/1laYUWYCBM6W4koSRNRHsbGx3HDDDezfv59rrrmGZ599luDgYNLS0ti4cSP/+c9/iI6O5qOPPqrW8fUmE4/ExDh/f6nZuHEjL7/8Mvfcc48kQeowi8XCddddR3R0NI8++ihdu3Zl8+bNTJ06lUOHDrFo0aILHuPw4cPk5+dzxx130KRJE+x2O9HR0bzyyissXryY6OhoTOf8G96+fTtdu3bl5ptvJiAggMTERL788ksGDRrEp59+yt13312bL/mSovdqelGfoQlbHqbRZZPxbXkT1oJkTK1NtJw8FktWFscXLMBacIzkdUsIGXpbDUZdu0Z0GMHO+J2sOLDCLUkQnQYa+2mJT7OV2K4AgV4adFJbLoS4hNSdb7LCZax5eQBovbwq/RxVVdl6wLE0Wp8OdecOb1JuElNWTQFg/OXjGdZumJsjEqL+MpvN3HjjjRw6dIjFixczZsyYEvunTJlCTEwMv/zyS7XPoWg0NO7c+WJDrbfy8vLwqsJ7tyjb//73P7Zv386cOXOYNGkSAA888AB+fn6888473H///fTr16/CY9x1113cddddJbY98sgjdOjQgWeffZalS5cybtw4576ypvhMmDCB1q1bM2PGDEmC1CCbOYXkP58jfe8CrAXJJfqHmHo4fs1K/gffhK6Ywjq6Kcqqub799UzfMJ2YpBiOZxynZUBLl55fURRu6W3irZ9yS2xXgVE9TXV2qp8QQpRF8rYNkO1MEqQqlSCxyTYSMmzotXBV27oxFcZmt/HUz0+RXpBOp8admDpoqrtDEqJe++STT9i3bx+TJk0qlQAp1qVLF+eXynPFxcVx6623EhAQgJeXF0OHDuXw4cOlxqWlpTFhwgSaN2+OwWCgSZMmPPDAAySct5KDqqrMnz+frl274uPjg7e3N61bt+aee+6hoKCgxNidO3cSFRVFUFAQRqOR9u3bM2PGDKzWko0Vi/uXVDbW8w0aNIiXX34ZgJYtHVPyFEXh008/LTEuIyODBx98kMaNG+Ph4UHfvn35888/S4zZuHGj87nvvvsunTp1wmg08sYbbzjHfPfdd/Tr1w8fHx88PT3p2bMn339fujn0zz//zMCBAwkODsbDw4MmTZowcuRI9u3bV2psZWIDyM/P54UXXqBt27YYjUaCg4O57bbbKvXnBI6/vzfeeIPWrVtjNBpp164db7/9dqWeWxO++uorTCYTjzzySIntkydPBuDLL7+s9rFbtnR8Oc3IKD2d43ze3t4EBQVVaqw4q+K+IQYCujyB1qMR1vzT5TZQVbR2EpZ9XZth1qhGno3oG9EXgJ8O/OSWGDqH62kScParg6JAi2AtncPrVqN8IYS4EKkEaYCsZ6bDVKUSpLgh6hWtDHga60bu7N3t7/LnqT/x1Huy4MYFGHV1p0JFiPpoyZIlADz88MNVel5eXh4DBw6kd+/evPbaaxw/fpy33nqLUaNGERMTg1arBSA7O5t+/fpx6NAh7r77bq666ipiYmL48MMP+eWXX4iOjiYkJASAV199lRdffJERI0bwn//8B61Wy4kTJ1i5ciV5eXnOaQirVq0iKiqKNm3aMHnyZAIDA/njjz948cUX+euvv5yvqaqxluW///0vgYGBLF26lHnz5hEUFARAnz59SowbNmwYjRs3Ztq0aaSmpjJ37lyGDx9ObGwsPj4+JcbOnz+f9PR0HnzwQUJCQggPDwfghRdeYMaMGQwbNoxXXnkFrVbL0qVLGTNmDO+88w6PPfYYAJs2bWLkyJFERkby3HPP4e/vT0JCAr/99huHDx+m83lVN5WJzWq1cv3117N582aioqKYOHEiJ06c4N1332Xt2rX88ccfdOxY8d31SZMmMX/+fHr37s0TTzxBZmYmr732Gk2aNKnweefKz88nPz//wgMBvV7v7NFht9vZtWsX3bp1KzFdBaBFixaEhYWxY8eOKseRn5/PX3/9xXPPPYder2fIkCFljk9NTcVut5OUlMQnn3zC/v37pQqkiirTNySw039I+/tNMg+VP7UpOyaGrL//xu+yy2oz3BozouMINsduZsXBFTze+3GXV18oisKYvp7OahBVhRE9pApECHEJUkW5srKyVEDNyspydyg1KnbhQnXXnXeqp5ctq9R4c5FdfWJhuvrAu2nq/lNFtRxd5fx+4ne19Rut1VZvtFKX7avc66hJBWl71cNftSj3UZC21+UxCVHbAgMDVR8fnyo9Z+DAgSqgvv766yW2z549WwXUNWvWOLf997//VQF1/vz5JcZ++eWXKqA++OCDzm3dunVTO3XqVOG5CwoK1MaNG6v9+/dXLRZLiX1z585VAfW3336rVqzlmTZtmgqox48fL7Xv7rvvVgH1kUceKbF98eLFKqB+8MEHzm2//fabCqiBgYFqSkpKifE7d+5UAfW5554rdY5Ro0apPj4+anZ2tqqqqvrUU0+pgJqcnFxh3FWJbeHChSqgTpw4scTYrVu3qoA6ZMiQEtsB9e6773b+fPDgQVVRFLVfv35qUdHZz5TY2FjV09Oz1N9LeYr/rCvzGDhwoPN5qampKqDeeuutZR63R48eanBw8AXPX2zy5MklztWhQwd11apVZY7NyckpMdZoNKr333+/mpOTU+nzicq70Gf1nkdHq3snTlRtZrO7Q62UbHO22nFeR7XVG63UmMQYt8Rgt9vVl7/LVB94N0194N00ddWuPLfEIYQQZans9/e6cUtfuFRxTxBdJStB9vxbREGRSiMfDe2burd4aNuJbQz+ZDCPr3gcFZUxXcYwqtMol8ehNQaAUvafhaIxOvYLUc9kZ2fj6+tb5edpNBomTJhQYts111wDwJEjR5zbli5dSmBgII8+WnL5y3HjxtGmTRuWLl3q3Obv709cXBxbt24t97zr1q0jOTmZu+66i8zMTFJTU52P4cOHA5TqX1LZWC/GU089Venj33XXXc6KkmJff/21c9+5ryk1NZWRI0eSk5PDH3/8AeBszrpkyZJS03+qG9vSpUtRFIUXXnihxNi+fftyzTXXsH79erKzs8s9x4oVK1BVlcmTJ6PXny2jj4iI4I477rhgjMXuuusu1q1bV6nHnDlznM8rrh4pbwUjDw+PSleYgKMyat26dSxZsoRJkybh7e1d7vQWk8nEunXrWL16NR988AG9evXCbDaXmsIlXEPn50dRaioJy5a5O5RK8TH6cE0rx//JlQdXuiUGR28QT/w8HdUfP+00k5ptu8CzhBCibpHpMA1Q8eow2kr2BNl60DEVpm8HIxo3lTzGZ8WTXpDO65teJzYjFoBwv3Ae6P6AW+LReYah8wzDmncK/06P4tP8eue+4lJcIeobX19fcnJyqvy8Jk2a4OHhUWJbo0aNAEcPkGL//vsvl19+eYkvxuC46O7cuTPLly93JmJmzpzJ6NGj6d+/P2FhYQwaNIjhw4czZswY55fbAwcOAPDggw/y4IMPlhlbUlJStWK9GK1atSrxc0XHb9u2baltxa+rU6dO5Z6j+HU9/vjjrFixgscee4znnnuOvn37MmzYMG6//Xbn1KKqxvbvv/8SEhLi3HeuyMhINmzYQGxsLF27di0ztmPHjgGUOWWmotdUVqznx1sZnp6eABQWlr3CV0FBgXNMZbRt29b593TLLbewdu1ahg0b5limfezYEmO1Wi3XXnut8+cHHniAQYMGcc0117B79+5S//ZF7Qq58UZOvb+cpNWrCezTB9OZ6WZ12YgOI1h9eDUrD67kmQHPoFFcfz+zU7ieN+72583lORw+beWbLfk8PtxbpsUIIS4ZkgRpgKpSCZKSbeNQvBUF6NPBfQ1RBywcUGrbqaxTDP10KMemHHN5PPmJ27DmnUKj86ZR50fR6GW1BlH/RUZGsmnTJo4dO0br1q0r/byK+mioqlqpY5w/rmfPnhw9epRffvmF3377jd9++41vvvmG6dOns2XLFkJCQpzPmTVrFldeeWWZxz2/B0VNxHoh5Z2jrOOX9WW8eNyqVavK/dJc3OsjMDCQHTt2sHXrVtatW8fmzZuZPHkyL774IsuXL2fQoEFVjq2iP4eq/Bld7Bem3NxccnNzLzwQMBgMBAYGAhAQEIDJZCIuLq7MsfHx8TRr1qzacQ0dOpSQkBDee++9UkmQ82m1Wu644w4eeeQRNm/ezODBg6t9XlFacQPVMpe01+jxv6w3Od3jydy5k5P/+x/t/u//UDR1u0h6UKtBeBu8ScxJZFf8Lno06+GWOBRF4c6BXrz8XRb/nLCw57iFK1rVjcb5QghxIZIEaYAutDpMeo6NHLPjQnZTjBmAiMZat2b45wyfw+RVk0ttnzt8rhuigawjjpUDfFrdJAkQ0WDccsstbNq0iYULFzJr1qwaP36rVq04fPgwFoul1Jf7/fv3ExQUVGI6jpeXF1FRUURFRQHw6aefcu+99/Lee+/x8ssv065dO8CRSDj37nttcsX7ZLt27VizZg3NmjUjMjLyguM1Gg0DBgxgwABHMvnAgQNceeWVTJs2jU2bNlX5/K1bt2b16tWkpaWVqgbZt28fGo2GFi1aVPh8cPydFv8dFdu/f3+l43jzzTedq/FcyMCBA51L1Go0Gq688kp27dpFQUFBieaosbGxJCQkMGLEiErHURaz2Ux6enqlxhZPhanseFF5ZTVQTY9ZQF7cOnSeYWg9gmg2fjzZMTHkHT1KzMSJWHNzMYaGEhYVRUAP9yQYKmLUGRnadig/7PuBFQdWuC0JAhAWoGVYNw9+3mXm2y35dGqmx8Mg1SBCiLqvbqe7Ra2wVbA6jMWmMuP7bF5d4nhsOVAEOJbInfF9FhZbzdwJrQpVVTmUcqjU9siQSEZ2HOnyeCx58eTF/wqAf9vxLj+/EO5y//3307FjR+bMmcOPP/5Y5pi9e/eW6L9QFVFRUaSnp/Phhx+W2P7tt99y9OhRbrrpJue21NTUUs8vrvYo/jI5dOhQGjduzOzZs8scX1BQUK3pPRXxPpNcrs0lT++8804Ann/++TL7fCQnJzt/X9brbteuHT4+PtX+0h0VFYWqqsycObPE9j/++IMNGzZw7bXXVtg7ZuTIkSiKwpw5c7BYLM7tJ06c4Kuvvqp0HNXtCQKOPjMFBQW8//77JbbPnetIrBf/GRc7ePCgcxpPscTExDLj+uyzz8jKyqJXr17ObRkZGRQVFZUam5eXxyeffIJGo+Gqq66q9GsXlaf3aopHYBfnI6TXG2g9grHmniRj37sYAgMJ6NkTAEtGBqrFgjkujuMLFpARHe3m6Ms2oqMjSbf68GosNssFRteu4VeaCPbVkJFnZ0W09LYRQlwapBKkgbEXFWE/cyFWVhJEp4FAbw05BTbOTXcoQKCXBp2L02aqqvLqb6/y6e5PAQjzCePRXo+y+J/FAFjsFgxa15ZfZh39GlQ7ppDeGPzauPTcQriTyWTip59+4oYbbuDmm2/m2muv5brrriMoKIi0tDQ2bdrEqlWryu2/cSHPPPMM3y9ZwoQJE/jl668Zdscd7D9wgA8//JBmzZoxffp059iOHTvSq1cvrrrqKpo2bUpSUhILFy5Ep9M5m2t6enry+eefM3r0aDp06MB9991H27ZtyczM5ODBg/z4448sXbq01JSQi9HzzJepqVOnMnbsWIxGIz179qRly5Y1do4ePXrw8ssvM23aNC6//HJuvfVWmjRpQkJCArt27WLVqlXOL9wPPvggcXFxXHfddURERFBYWMiSJUtITk7m6aefrtb577nnHr744gvmzJlDbGws11xzjXOJXF9fX+bPn1/h89u3b8/EiROZN28eAwcO5LbbbiMrK4v333+fDh06sHv37krFUd2eIODoxfHpp5/yzDPPEBsby2WXXcbmzZv5/PPPGT9+PP379y8xvmPHjkRERBAbG+vcNnz4cBo1akTv3r1p3rw5WVlZbN26leXLl9OsWTNeeukl59hNmzbx8MMPc/PNN9OmTRt8fHw4fvw4X3zxBXFxcUybNo2IiIhqvRZRNVqDH8HdXyZx66Ok73sf7+bDyfv335KDVBUUhcRly+pkNUjv5r1p5NmItPw0tp3YxqBWg9wWi0GnMG6AY9nc9f+Y6dXeQPMg+XohhKjb5F2qgbEVd7xXFLTnlAAXUxSFUT1NzjXgi6nAqJ6uXQvertqZ9us0vv7bsRLCtGumMb7beEezua5j3ZIAsdsKyT72HQB+UgUiGqBWrVqxa9cuFi5cyPfff8/MmTPJyckhICCAK664goULF5a6i15Zvr6+/LZuHaMjItj0xx+s2bmTRo0acffddzN9+vQSjTwnT57MqlWrePvtt8nMzKRx48ZcddVVfP311yXuwA8dOpTo6GhmzZrFV199RUpKCgEBAbRu3ZpJkyaV27yzuvr168eMGTP48MMPuf/++7HZbCxatKhGkyAAL774IldeeSULFixg/vz55OXl0bhxY7p06cJbb73lHDd+/Hg+/fRTPvvsM1JSUvD19aVDhw58/fXXF+xXUR6dTsfq1auZMWMG3333HStWrMDX15cbbriB6dOn0759+wseY86cOYSFhfH+++/zzDPPEBERwdSpU/H19eXee++tVlxVodfr+eWXX5g2bRpLlizhww8/pHnz5syYMYNnnnmmUsd44IEH+OGHH/j4449JTU1Fr9fTunVrnn32WaZMmVJiqlBkZCQ33ngjv/32G1999RX5+fk0atSIHj168MEHH3DDDTfU1ksVZfAOH4ZXs+vIi/uF5D+nUpjoX3qQqmJOSHB5bJWh0+gY3n44X+z5gpUHV7o1CQLQpbmB7q0N7DxWxJeb8njuJl+3NdIXQojKUNSa6vRWD2VnZ+Pn50dWVla1loWsiwri4jgwdSpab28uO68MuJiqOqbEnEhxLHmmABHBWp6/xddlSRCb3cbzvzzP9zHfo6Awa+gsbom8xSXnrkh27HKSfp+I1hRCy1FbUTSSRxSiJhXl5THzzJSSqbm5GCq5lLcQQlSFNT+REz9fh92Sgy3hCvL/9nJUgJzD1Lw5HWfMcFOEFdt9ejdjvh6Dl96LPx/9E5O+9I0tV8rMs/N/X2ditsAdAzwZ1MXjwk8SQogaVtnv79ITpIG5UFNUcFSDDOxsdP7s6ioQq93KlNVT+D7me7SKlrk3zK0TCRA42xDVr804SYAIIYQQlyidZyhBlz8HgLbJPhSPAjjvOsfUvLk7QquUbmHdaObbjDxLHr/9+5u7w8HfS8Pono7VrL7/PZ/9p4o4kWJ1PtJz7W6OUAghzpJvcQ1M8fK4ZfUDOZf9nLshLYK1dA4vexnGmrTtxDZeXv8yAaYAdsbvRKfR8daNbzGs3bBaP3dlFGbsx5yyExQdfm1ud3c4QgghhLgIvm1uJ+fEcgqSd+A/LAvzvkYUpaah8fDAlptLxs5f8WrdmmAXrS5VFYqicGOHG/lgxwesOLCC4e2Huzsk+nU0sHhbPoVWmLey5LRqX0+FWeP90WtlmowQwv0kCdLAFK8Mo7tAEuTIacdUGG8PhahenrVaBRKfFU96QTqzNs3iWLqj+75Oo+OVIa/UmQQIQObhLwDwDh+KztTYzdEIIYQQ4mIoiobAyCeJX38HVvMRdK2PoGt9dr9qU4j77kM0Hh406tfPfYGWY2THkXyw4wM2Hd9EtjkbXw/3Tt026BRC/DUkZJSs+nBXc30hhCiPvB01MJWpBFFVlcOnHcsuPnSdN51quQpkwMIBjP5yNPuT95+N025l6tqptXreqrAVZZMTuxwA/3bSEFUIIYSoDzT68hMHilZFMVg58dFHdXK53PbB7WnbqC1FtiKGLhrKthPb3BqPoijc2tez1HZ3NNcXQoiKSBKkgXFWglTQEyQ1205Gnh2tBlqF1H6x0OO9Hy9z+9zhc2v93JWV/e8PqLYCDH7t8Qi+yt3hCCGEEMIF/Lp1A1Ul9t13yfrnH3eHU0J8VjxXhTuuSZLzkpm9eTZ7E/cSnxXvtpg6h+sJD9KW2BbeyDXTqoUQorJkOkwDU5lKkOIqkBaNdRj1tZe1V1WVj3d+zHt/vFdqX2RIJCM7jqy1c1eFqtrJOuKYCuPX7k65kyGEEEI0EKGjR6MWbCJzxw6OzZ2LoVEjLBkZGENDCYuKIqBHjxLjM6KjSVi6lMLExHLH1JQBCweU+DkmKYbRX44G4NiUY7VyzgtRFIWbepl466ezPUEy8+0kZdkJ9ddW8EwhhHAdqQRpYCqzOsyh0xYA2jWpvRxZljmL/yz7D7M2zcKOHT8PPzo37swrQ14hMiQSAIvdUmvnr4qCpN+x5BxHo/PGt8Vod4cjhBBCCBdRNBpaPPIIphYtwGajKDkZ1WLBHBfH8QULyNixA1VVUVWVjB07OL5gAea4uJJjamkqTXkVs7OHza6V81VW53A9LYIdCQ+dBnIKVF7/MZt/k6xujUsIIYpJJUgDU5VKkNpKgvyT8A9PrHyCuOw4DFoD/3fN/3Fz55sxaA0oisLYrmOx2C0YtIZaOX9lWPLisRVmAJAW8y4AnmEDsBVlodGXn0ASQgghRP2hqioanQ7Vaj1/BwDH3367rCed/VVRSFy2rFaqQUZ2HMmiXYvYm7S3xPbPdn/GFU2uoGVgyxo/Z2UoiqOp/rdb8xjZw8SaPWZOpNiYszybh67z5rIW7ru+E0IIkCRIg3Oh1WHScmyk5djRKNAmtObmbxYvf9snog/f/fMdRbYimvs15+2Rb9MlpEuJsYqiuD0BcmLlYFR7YYntuadWkRe/nogR69F7NXVTdELUbwYvL6ads0S3EELUJq0xAEVjLPWZXyz72Dd4BHahMCmpeidQVcwJCRcRYfmKK2YjQyK5teutfBz9MXFZcexL3seIz0cwbfA0bulyi1um8XYK1zN9rD8AXSIMfLg2l5iTFt5bncudAz3p38nD5TEJIUQxSYI0MBeqBCmuAokI1uJhuPgPzeLlb2dunMmx9GPOJXD7t+jP2yPexsfoc9HnqGm2woxyL4ZUeyG2wgxJggghhBD1gN6rKREj1jurP4vlnlxNxv73yD76DVq9L8bQEMxx8WerPAAUBY8mTWj7/PMAHHntNcynT5ccA+j9/WsldoPWwOJxi9Fr9M5K2risOJ5b+xzbT23nubXPsen4JmZcNwM/D79aiaEyPPQKj13vzReb8vj9YBGfb8wnNtlG/06GEgkaH5OGQG+ZqS+EqH2SBGlgLrQ6zNmpMDVTBXJ+065iW2K31MkEiBBCCCEaFr1X01I3NzwCu6DzDCVl54tkHPgQ3/6jMX9tB0XjnOaCqtLkllvQ+zqW2Q27+WaOL1jg3FesKCODnIMH8enQocZjP7dyVlEUwv3D+XzM53wc/TFzt81l9eHV7EnYw71X3MuSvUt4cfCL9I3oW+NxXIhOq3DP1V74mBTW7ilk837H41y+ngqzxvuj10oDeiFE7ZJ0awOi2u3Y8vOBiipBarYpanlNu+rS8rdCCCGEEOfzbzee4CtfAqAgYxlBtxkwtfVEG2jG1NaL8IdH4NUxzDneu1MTwh+6EVNbL7SBZjzaemHq4IOiy+XYnDnk/fuvS+LWarQ83PNhloxdQjPfZiTmJDJz00yOph/l9U2vu20ZXUVRuLmXJ4HepZMcChDopUEn30yEEC4glSANSHECBMruCZKZZyc5y46iQJuwmvmn4e/hX2pbXVr+tiyqzezuEIQQQghRB/i3vxtrYQYZMW9RmP0butaga+3Yl3lyDVlxRiJGrAdw9hM7dwyAV4SWvE3dODp7Nu3++19M4eEuib1rWFfisuNKbNuXvM+ty+gqisL4QV4lltAFUIFRPU1u6V8ihGh4JN/agBT3A9F4eKDoSic5ipfGDQ/S4mm8+H8aeUV5vLDuBQAaeTaqk8vfnk+120j96w13hyGEEEKIOsK72bXl7lPthWT/u4Tsf5eU209M0dgwtQrFlpfHkddfr7VGqWUpr/L2oR4PuSyG8xUvoXtuusOog1Yhcm9WCOEakgRpQC60Mszh+DP9QGqoCmT+tvmczjlNE58mbLh/A+MuG8fSO5eyeNxit67+Uh5VVUnZ9RLmlB3ljlE0RrTGABdGJYQQQoi6LH3vW6TvfavCMeF33YWpeXOsWVkcef11ClNTXRLbyI4jnTegzvVR9EfM3ToXm93mkjjOpSgKo3qaOLd9bKEV5q/MIc9sd3k8QoiGR1KuDYhzZZhym6IW9wO5+Kaofyf8zae7PwXg1etexdvoOKe7l7+tSMaBD8k68iWgENz9FTyCLis1RmsMkJVhhBBCCOHkEdQdAHPqznLHaE0m2jzzDIdnzKAwIYFDL72EzsuLwpQUjKGhhEVFEdCjR43Hdv4yut/98x2ns0+TXpDOu9vfZVf8LubfOJ9gr+AaP3dFiqtBYlNshAVoyMqzczzZxtwVOTw10gdvD7lPK4SoPfIO04BUVAmSnW8nMdORfW97kU1RLTYLU9dOxa7aGd1pNANbDryo47lCduxy0v56HYCgK17Av90deAR2KfWQBIgQQgghzhXcfRrB3adVOEa1WdD7+dH2uefQ+fhgzcrCfPo0qsWCOS6O4wsWkBEdXeOxFS+ju/TOpYy7bBzL7lzGtv9sY94N8/DUe7L91HZGfD6C7Se31/i5K6IoClG9PAkL0HB7Py+ejvLFx6RwMtXGm8tyyM6XihAhRO2RJEgDYiuuBCkjCVK8NG7TQO1FZ98/iv6IQ6mHCDQF8t9B/72oY9UGS1485vQY5yPzyDck/TEFAP/29xHQ4T43RyiEEEKI+iT17zdQ7RYMgYGlK3LPLLmbuGxZrZzboDU4G44WV+SO7DiSZXcuo22jtqTkpTB+yXje2/4eW2K3MPR/Q9l2YlutxHKuTuF6po/1p1O4nmaNdDw9yhc/T4X4dBtvLMsmM08SIUKI2iHTYRqQ4ukwujKmwxxOODMVpunF/ZM4lnaMd/54B4AXrn6BQM/AizpeTbPkxTu7t5emwa/9vS6PSQhxlrWwkLWTJgEwdO5cdEajmyMSQjR0WmMAisZY5rXDub3CyhsDYE7+g8RtEwjtu4CisvqBqKpLG6YCtG7Umh/v+JFp66fx474fmbN1Dt4Gb3KLcpm9eTavDnmVQFMgTf1cUwUbFqjl6dG+zFmeQ2KmnVk/ZnFHfy98vUrenPMxaQj0dmxLz7GRY1ZLHevcMUIIcT5JgjQgxdNhyqwEOdMUtf1F9AOxq3b+u+6/FNmKGNBiQJ1cBtdWmFHuBQrYsRdlAs1cGJEQ4lx2q5Wd770HwJDZs0GSIEIIN9N7NSVixHpshRml9p3bK6y8MYWZh0jZ8Ty5p9aQ+PtEjKGNMceddlSAnMMYGlo7L6ACngZP3rj+DX7c9yMAuUWOa8WYpBi3LKUb4q/l6Sgf3lyWTVqOyoJVuaXG+HoqzBrvD8CM77PJLiidBCkeo9fKkrtCiNIkCdKAWMuZDpNrthOf7ugO3vYiVob59p9viY6LxlPvyatDXpW13oUQQghRL+i9ml6wL1h5YzwCu6A1+pOw5RFyT67C86orKcq2ApoSiRCNqRDVakXRuf7yfO7wuUxaNanM7a4W7Kvl6dE+/N832VjLWLxGp4Gt+82O32tL71eAQC8NOikEEUKUQ5IgDYi1uDHqedNhjpzpBxIWoMHXs3qfGIk5iby+ydFYdHL/yS4rnRRCCCGEqOu8mw4mrN+7JGx5FEv+Lrz6lB6j2v7m+MeNafnQ0yga136DH9lxJIt2LWJv0l7ntsiQSLdV9Qb56rjnai8+/jWv1L70XJWvtxSU+1wVGNXTJDfjhBDlkhxpA1JeY9SLXRpXVVWmrZ9GblEul4ddzvjLx19coEIIIYQQ9Yx3syEEXf5sufsVrUr23u2c+uwzVLX0FI/aVLyUbsuAlgBoFA121e7c7g5XtTUQ3qhkqYenAbq11HFFKz1XtNLTraUOT0PJ54X6a+gcXv3p3UKI+k8qQRoQWzmNUYtXhmlXzaVx52+bz69Hf0WjaHjtutfQasqoTawjVFuRu0MQQgghRANlCulV8QBFIXXDBjRGI03HjnVZNUPxUrpaRcvQRUM5nnGcmzrfhEFruPCTa4miKNzU28RbP53tC/Lgdd50aV4yppiTRSXGZOTZiU+z0SxIvuYIIcomlSANSFk9QfIL7ZxKdUy4rGolSHxWPH+c/IP3/3wfgEaejSiyFRGfFV9DEde8nNgfy913bod3IYQQQghXCx3pmH6SvHo1iUuXuvTcBq0BrUbL+G6Oit5v/v7G5RUp5+scrqdFsOPmWotgbZkVHueOMeqg0ALzVuaQnFVGQxEhhEAqQRoMVVWdq8PozkmCHEmwogKN/TT4e1UtJzZg4YASP6fkpbilk3hlmdP+JuvoNwAEdftvqbsx53Z4F0IIIYRwNf/u3dGoYcR9+SUJS5eS8ttv2PLyMIaGEhYVRUCPHrUew02db2LOljkcTT/K7yd/p29E31o/Z3kURSGqlyffbs0jqpdnmZUxJcb0NLEi2kxcmo25K3J4NsqXAFkqVwhxHnlXaCDshYWoNkdG/NxKkOKpMNVZGnfO8DllbndHJ/ELsdsKSdr+DKh2vCNGEtDxATwCu5R4SAJECCGEEO6kqnYaDx2Kfy/HjRprZiaqxYI5Lo7jCxaQER1d6zH4GH24qfNNAHyx54taP9+FdArXM32sP50q6PNRPKZbKyNPjfChsZ+GtBw781Zmk1Ngd2G0QohLgSRBGojiKhBFp0NjNDq3n22KWvWioHC/8FLb3NlJvCIZMe9QlHUYrbERja+c5u5whBBCCNEAaY0BKBpjufuzDn2KqqqY48+bWqyqoCgkLltWuwGecWe3OwFYf2x9nZ7mXBZfTw2TRvrg76WQkGFnwc85mIvcO61HCFG3yHSYBuLcfiDFpYTmIpWTKcX9QKr+T2HpPsdcVX8Pfyb3n8zifxYDjg7j7mykdT5z+j7S9zv6lgT3eBmtR6CbIxJClEdRFII7dXL+Xggh6hO9V1MiRqzHVphRYnte/AbS984jJ3YpGr03hYkJpZ+sqpgTytheC9o0akPfiL5sO7GNr/7+imcGPOOS89aURj5aJo3wZfaybGKTbby5PJux/T3Rac9+rviYNATKVBkhGiRJgjQQZa0MczTRgl2FIF8NgT5VW9GlyFbEmiNrAJh3wzwGtBzA2K5j61wCRLVbSP7zGVBteIcPw6f5De4OSQhRAb2nJ4/u2+fuMIQQotbovZqWmoLrmJbbhKTtz5B15AtMV7TEfNAPzi1gUBSMQc1dFuf4buPZdmIb3/3zHRN6T8BD7+Gyc9eEsEAtj1/vzaylOZxIsTHrx5wS+309FWaN90evlYS7EA2NJEEaiLJWhjkcX/2lcbfEbiGjIINgr2D6RPQBHHdt61ICBCB93/sUZuxHYwwguPt0d4cjhBBCCFEm31a3oKp2kv98Fm3gcbz6lDFIicGS97RL+phd0+oamvo2JT47np8O/sQtkbfU+jlrWqtQHSH+GpIyS/YFUYBALw06KQQRokGS//oNRFkrwxQ3Ra3q0rgAy/YvA2BEhxHoNHUnl2bJi8ecHoM5PYacEz+RHvM2AIFdJqAzBbs5OiGEEEKI8vm1vpWAzo+XP0C1lJpKU1u0Gi13XH4HAJ/v+dzty+VWh6Io3N7Ps9R2FRjV0yTTLoVooOpkEuSbb77hyiuvxGQyERQUxNixYzlx4sQFn5eRkcFbb73FddddR3h4OCaTifbt2/PQQw9x6tQpF0RedxVXguR5BnEixcrRBAvHkx1JEC+jQnpu5Ttn5xTmsP7YegBGdRpV88FWkyUvnhMrB3NqzQhOrRlB4rYnQHW8xrQ9s7DkXVqNvYQQQgjR8HiHD3V3CE63Rt6KUWdkX/I+dp/e7e5wqqVzuJ4WwVrOTXcE+2roXMFqM0KI+q3OJUHeeecdxo0bh8lkYt68eUycOJF169bRp08fTp8+XeFz//zzTyZNmoSqqjz22GO8/fbbDB8+nC+//JLIyEj279/voldR99hyc7EpWv6nDuHVJdm8vjQH+5mE/rurc5nxfRYWW+Uy/GsPr6XQWkibwDZ0bty5FqOuGlthBqq9sMx9qr3QZXdOhBBCCCFqi2q1uuxcAaYARnV03PD6fM/nLjtvTVIUhVE9TSXaq2Tk2jmVanNbTEII96o78xiAtLQ0pk6dyhVXXMHGjRvR6RzhDRs2jKuuuooXX3yRjz/+uNznd+jQgUOHDtGmTZsS22+44QaGDBnCtGnTWLJkSa2+hrrKmpeHRrXhpysk32Is2WeLqs2LXHZgGeCoApEyQiGEEEII18n66y9M113usvON7zaexXsXs+bwGpJykwjxDnHZuWtKcTVIbIoNDz2YLY6bgP+9xRdfzzp3T1gIUcvq1P/65cuXk5uby4QJE5wJEIDu3bszYMAAFi9eTFFRUbnPb9GiRakECMC1115LYGAge/furZW4LwW2vDwU4LrGSZxf71GVeZEJOQlsP7kdgJEdR9Z4nEIIIYQQonzpmzej2lxXxdCpcSe6N+2O1W7l27+/ddl5a5KiKET18iQsQMN9g70I8deQnmvnvTW5la6EFkLUH3UqCbJjxw4A+vQp3Q67T58+5OTkcPDgwSofNysri5ycHBo3blzhuMLCQrKzs0s86gvrmcaoHYLtRASfXQ5XUaBFsLbS8yJXHFiBikqPZj1o5tesVmKtrqKsw+4OQQghhBDiomiNASgaY5n7VLuGwqQcMs5cM7vK+G7jAfjmn28ospV/Q7Iu6xSuZ/pYf7q1MvL49T6YDArHEq18tSnvkmz6KoSovjqVBImPdzSubNas9Jfr4m1xcXFVPu6rr76KxWLh7rvvrnDczJkz8fPzcz7Cw8OrfK66ynamMare24vrLj+7zruqVq079vL9ywGc80PrioKUXSRH/5+7wxBCCCGEuCh6r6ZEjFhP+LCVhA9bSWi/d537PEz3o5o9SFq50qVf3Ie2HUpjr8ak5KWw5vAal523toQGaHnoOi8UBbYdLGL9P2X3lBNC1E91KgmSn58PgNFYOvvt4eFRYkxlLV68mDlz5jBkyBDuvffeCsdOnTqVrKws56M+rShTvDqM1tubIJ+zf+1VqQI5mHKQQ6mHMGgNDG8/vFbirI78pO3E/3YXqjUfKDuZo2iMaI0Brg1MCCGEEKIa9F5N8QjsgkdgF3yaD8c7wjEFWRtwFI2HBwWnTpH999+ui0erZ9xl44BLt0Hq+bo0NzCmj2P53MW/57PvpMXNEQkhXKVONUb19HS8ERUWFmIymUrsKygoKDGmMlatWsX48ePp1q0bS5YsQaOpOOdjNBrLTMDUB7Yz02F0Xl5k5DruHOi1ENXLs9JVIMv2LwPg6lZX4+fhVytxVlVewhYSNj+EajNjCu1L8JUvo9oKSo3TGgPQezV1Q4RCiKqwFBTw1fXXA3DH6tXoz/ssEEKIhqhRlwnknvyJ/MTfCLz6MVJX/03iihX4XnaZy5rU337Z7by7/V32nN7DwI8G8trQ1+gb0dcl564t13Y1Ep9mZdvBIj5Ym8O9g71pdM7NQh+ThkDvOnXPWAhRA+pUEqRpU8eX1Li4ONq2bVtiX0VTZcqyZs0abrrpJjp06MAvv/yCn1/d+NLuDnarFXuho8xP6+1NWoKjmdblLQ10qmQViM1uY+WBlYBjVRhXs+TFl1ritiA5mtQ9M0G14NnkasL6v49GWz+TWEI0FKrdzolNm5y/F0IIAQa/1vhEjCQndhmaxntR9Hryjhwh99AhfDp0cEkMRdYiejfvzebYzcRlxzF782xeHfIqgaZAmvpdmjeaFEXh1n6ebD9chNkC76/JLbHf11Nh1nh/9FpZDVGI+qROJUF69OjBhx9+yO+//14qCfL777/j7e1Nh0q80a9du5aoqCjatWvH+vXradSoUW2FfEko7geCoqA1mUjPcVRKBPpUPrP9Z9yfJOYm4mv0ZVDLQTUfZAUsefGcWDkY1V72fE1T6ACa9P8ARWtwaVxCCCGEEK4S2OUJck6soCB5MwED7id9/SGSVqxwWRJkwMIBJX6OSYph9JejATg25ZhLYqgNJr1CkwAtp9JKrrijAIFeGnRSCCJEvVOn/luPGjUKT09PFixYgNVqdW7fuXMnmzdv5tZbb8VgcHzRTUhI4ODBg6V6hPzyyy+MHj2atm3bsmHDBoKCglz6Guqi4pVhtJ6eKBoNabmOu6uNqlDeV9wQdXj74Rh1rq22sBVmlJsAAWjU9SlJgAghhBCiXjP4tsKnxWgAtGH7QVHI3ruX/OPHXXL+ucPnVmn7pUJRFG7qXXrqpUrVFg8QQlw66lQSJCgoiNdee43du3czaNAgPvzwQ2bMmMGwYcMICQlh+vTpzrFTp06lY8eOzmV1wZEsGTVqFKqqct9997FmzRq+/PLLEo+GqLgSROflBUD6mSRIZStBzBazsxP46E6jaz7Ai6Ro6lRBkxBCCCFErQjs8gQoWsxpf+DfLwKAxJUrXXLukR1HEhkSWWJbc7/mjOw40iXnr02dw/W0CNaWaK8f6q+p9OIBQohLS5379vjkk08SFBTEnDlzmDhxIp6engwZMoSZM2c6e4aUJyYmBrPZDMBTTz1V5pg777yzxmOu65yVIN7eAKTnVK0SZP2x9eQW5dLUtylXNr2ydoIUQgghhBAVMvi0wLflTWT/uwRd+GHAQGZ0NHvuvRdjWBhhUVEE9OhRK+e22B2rp0SGRBLsFcyGfzdwOuc0GQUZBHoG1so5XUVRFEb1NPHWT2d7ghQUqRRZwSh5ECHqnTpVCVLsjjvuYPfu3RQUFJCW4PFv4QAAQRBJREFUlsa3335Ly5YtS4z59NNPUVWVQYMGObfdc889qKpa4aMhOrcSpNCikmt2/DlUthJk+QHHVJiRHUeiUVz7T0ZVVXJPrXXpOYUQQggh6qrALo+DoqMwYwca/2wAVKsVc1wcxxcsICM6ulbOa9AaWDxuMUvvXMo7I9+huV9zrHYr7//5fq2cz9WKq0EAtBrIyldZ8nv+BZ4lhLgU1ckkiKhZxUkQrbc3GWemwnjowWS48BzH9Px0Nh13rNQwqqNrV4Wx5CdweuO9ZOx7x6XnFUIIIYSoq/TezfFtdRMAxrYnz+5QVVAUEpctq7VzG7QGFEXBqDMyfYhjmvpnuz/jQPKBWjunqyiKQlQvT8ICNIzu6egRsmlfIXtPFLk5MiFETZMkSANgPacS5Gw/EG2lGj2tOrQKq91K55DOtA1qe8Hx1WHJi8ecHuN8FKTtJW3vW5z4aQj5CZtA0YOiLfO5isaI1hhQK3EJIYQQQtRFgZ2fQLUr6IIy0QZknd2hqhTEx7tkifH+Lfpzfbvrsak2Xvz1Rezqpb+seadwPdPH+jOsm4lruzoWAvh0Qx45BZf+axNCnFXneoKImmcr7gni5UXamX4ggZXoB7LtxDZe2/QaAKM7jq6V2C60/K3BrwNh/d5G0ZmwFWaU2q81BqD3ujTXphdCCCGEqA69dzPs2eFo/U9iaH+cwv1tnPvUIh0HX3yRprffjm+XLrUax3+v/i+bjm9i9+nd/BDzA2Mix9Tq+Vwpqpcn+05ZSciw8cXGPB4Z5i0rxQhRT0gSpAEorgTRenmRnutYA72iJEh8VjzpBem8suEVCq2O5ESbRm2Iz4qnqV/NJhwutPxt456zMPg5Ptgl2SFE/afRaul0yy3O3wshhCjNkheP1v80ALqAXHR9/3LuU20a8jbD0ddfx7drV3y6dCFtyxYKExMxhobWaPPUMJ8wnuzzJDM3zeT1za9zbZtrCTDVjwpdg07hgWu9eO2HbPYct/DHoSL6dDC6OywhRA2Q6TANQHEliM7bu1LL4w5YOIDRX47mSNoR57Z7f7iXAQsH1Hhsqt1W4X5FI1+ChGhIdB4ejFmyhDFLlqDz8HB3OEIIUSc5qmOtZe5TtHYCB3QHrZbsf/4h/uuvMZ86hWqx1Erz1LuvuJt2Qe3IKMjgzS1v1thx64LmwTpG9HD0B/lmSx6p2RVftwohLg2SBGkASlSCVGJ53LnD51Zpe0XO7/dR/CjKjSPn5M8k/v5ElY8phBBCCCHKF3LDDXR6/XU0JlPJHbXQPFWv1fPKta8A8N0/37Hn9J4aO3ZdMKybB61DdZgtsGhDHnZ7w1xtUoj6RKbDNADOJXIrWQkysuNIFkYv5EDK2U7fkSGRjOw4skrnrbjfhwLIh4gQQgghRE0rSN6Bf/t7QZuLxreg1H5z6lEs2dnofX1r5Hzdm3Xnps438eO+H5n26zR+vPNHdJr68TVDq1G4b7AX07/L4vBpK4u35dP7vGkxPiaNc6p5eo6NHHPpa9yqjhFC1J768e4kKlRcCaJ4nq0EqegN1mK3kFOYAzjmegZ5Bjm3G7SGSp+34n4fKorOE5+IUWQf+6bSxxRCCCGEEBVL3f0K2ceW4Nn3MIqm9Momqk1h/9RHCRowgpDhw9H5+Fz0OZ8d8Cy/Hv2Vfcn7+Oqvr7j7irsv+ph1RWM/Lbf0MfHV5gLW7y1k/d6S17e+ngqzxvsDMOP7bLILSic4qjJGr5UGrELUJkk11nOq3e6sBCnQemK1g6KAv1f5f/UGrYFuTboBMLrTaJbeuZTF4xZXKQFSGU0GfkJgl8dQNGU3mZLlb4UQQgghqkFjpCjrYJkJEABFq6KSR9JPPxEzaRKnlywhbcsW9j//PHvuu4/9zz9f5b4hQV5BPN3/aQBmb57N4I8Hs+3Etot+KXXFgE5GTGVcCitAoJcGnQZ0GseNxvNTGFUdI4SoXVIJUs/ZCgoc8z+BLLsJyMffU4OuggyzXbU7P7T6t+iPoig1ngAB0Oi90Xs1JWLEeln+VgghhBCikrTGABSNscyKW0VjpNl135Nx8BNyY5eVe4ymd9xB6qq/KDhxgsQVK0rsK26gyoQJVVpJpl+LfrRp1IajaUeJzYxl9ubZvDrkVQJNgTW+wqCraTQa7hzoxcJ1eSW2q0Bipo1JizIBsNrUUhO+KztmVE+TLMMrhAtIEqSeK14ZRmM0kml2pJYr6gcCcCD5AOkF6XjpvZwVIdU6dxmJjbLovZpKskMIIYQQopIqcxMpoMP9FSZBfDp0IKjXTWTt3s3x995DtVjO7jyngWpVkiBXf3x1iZ9jkmIY/eVoAI5NOVbp49RVPdoYWL6jgOSskhU2ZgtgqbjX3YXGtAjW0jlcXwNRCiEuRJIg9ZxzZRhvb5JzHMt6Xajh0pbYLQD0at6r2hUgVnMqydH/rdZzhRBCCCFExS72JlLyjucJvuIF/LtfhWI0o5hKN1AtTD9epWPOHT6XSasmldr++tDXqx1nXaIoCmP7e/LWT7nObeMHetImrGTy4miChS825VdpzKAuRqkCEcJFJAlSzzlXhvHyqtTKMACbYzcDjqkw1WE1pxG//k6suafKHSP9PoQQQggh3KcwfS9xv96GMag7nn13l91A1a7Bkhdf6WTLyI4jWbRrEXuT9pbY/tVfXzGw1UCCvYJrJHZ36hyup0WwltgUGy2CtfTvVDp5ERagYcv+wkqPAYg5aaFvRw+XvQ4hGjJpvVPPWc9Mh9F6nV0ZplEFlSB5RXnsjt8NOOZ1VpXNnE78hjspyjqE1tSYpoO/IXzYylKPiBHrZQqMEEIIIUQtKe4bUiaNAe+IUaDRU5i6s/wGqhp7pac3g2MlQYDIkEheGfIKrQJaoVW0/JP0D1FfRnEw5WCVX0ddoygKUb08CQvQENXLs8zqjaqMCfJ1XJfvPGbhRIq11uMXQkglSL1X1UqQP0/9icVuIdwvnBb+Lap2rsIM4jbcSVHmQbQewTQb/DUG39bVjl0I0fAU5eXxVosWADwZG4vBy8u9AQkhxCWqMn1DLLlTSNk9nby4dTVyToPWwOJxi9Fr9I6pI13HciTtCI8uf5TjGccZ8/UY5t84n8GtB9fI+dylU7ie6WP9a2TMzDv9Wbgulx1Hili6vYCJIy5+uWIhRMWkEqSec1aCeHufTYJUUAly7lSYC81LtOTFY06PwZweQ37iH5z65RaKMg+gMQZIAkQIUW35qankp6a6OwwhhLjk6b2a4hHYpdSjuBpX792MwC4TKjxGiYaplWDQGpzXkIqi0C6oHT/c8QO9m/cm35LPw0sf5pOdn6CqFTcSbUhGX2VCq4F9pywcjK/an7cQouokCVLPFVeCqJ4+5BQ4PmwqSoJsjd0KXHgqjCUvnhMrB3NqzQhOrRlB/IZxWHL+BcBuyUXRmWoifCGEEEII4UapW7Zc9DH8PPxYdPMibu96Oyoqr218jft/vJ/r/ncd205sq4EoL23BfloGdHJMXfrxj3xJEAlRyyQJUs8Vrw6T6xEIgFEPnsayKzzisuI4nnEcraKlT/M+FR7XVphR5tr0ANgtVZo/KoQQQggh6qb0zZsxnz590cfRa/W8OuRVHu/9OACbjm/iWPoxZm6cyd7EvcRnxV/0OS5lN3Y3YdTB8WQbe/6VahAhapMkQeq54kqQXL0fAIHe2nKnuRRPhenWpBs+RpmPKIQQQghR31XYQFXVYC9QOLloUY1UJyiKwjt/vFNi24GUA4z+cjQDFg646ONfynw9NQy53LE6zNI/87HZpRpEiNoiSZB6rrgnSLbGG4BGFTRFrexUGCGEEEIIUT8UN1A9dxU/74hRAGgMfqAxkXvwIOk1MC0GYO7wuWVun9xvco0c/1J23eUmvD0UEjPt/H6wnIprIcRFkyRIPVdcCZKlegLl9wOx2q38fvJ3AAa0uHAmXuYqCiGEEELUD+c3UA3pOQu9byvslgz8BxcAEPfNN1hzci76XCM7jiQyJLLU9g/+/MB5Q66hMhkUhl/p6Ku3IrqAIqtcbwtRGyQJUs8VV4Jk2hzldeUtj/t3wt/kFObg7+FPl5AuFzxuQWLD/pASQgghhKivNDoPQnu9CYoGq3Unps5gy80l7uuvL/rYFruj30VkSCSvDHmFzo0746X3Is+Sx/0/3s+P+3686HNcygZ1NhLorSEzT+W3vWZ3hyNEvSRJkHpMVVVnJUhmkQ6ARuVUgmyJdZQ49onog1ajrfC4dmsBmYc+LXe/ojGiNQZUI2IhhBBCCFEXeAR1I6DTfwAwtNqLYrSQvnUrOfv3X9RxDVoDi8ctZumdSxl32TiWj1/Otv9sY0SHEVjtVp5e/TTvbn+3wVYd63UKI69yVIOs2m0mv9Du5oiEqH907g5A1B61qAjVagUgw6wB1HIrQYqTIJWZCpNxYCE2czJaj8aE9X8PRVuymZbWGOBcf14IIYQQQlyaArtMIC9+A0WZB/EdlEXW2kYcfeMNAIxhYYRFRRHQo0eVj2vQGpy/VxQFH6MPc2+YS5hPGB9Ff8TcrXNJyEngpcEvodM0vK8rvdsZWL2rgKQsO99ty+eaSA/nPh+TpsT09vQcGznm0gmjc8e5ckxNH0uI2tDw3lUakOLlcVWtjow8x5tMWW8qmQWZ/JP4D3DhpqiWvNNk7H8fgOAr/w9T8JU1GbIQooHT6vUMnDbN+XshhBDuo9EaCe09h5NrRmHXHETXpB3W040BMMfFcXzBApgwoVqJkFLnUjQ8O/BZwnzCmL5hOt/8/Q1JOUnc3vV2Zm+ezYuDX6RvRN8yn7vtxDamr59e4ZhLiU2F7ALHtfvvB4v4/WCRc5+vp8Ks8f7otQoWm8qM77OdY89VPA5w2ZiajEmvLXs1SyFqgiRB6rHifiAW32CsNlAAf6/SSZDfT/6OXbXTtlFbwnzCKjxm6l+zUG1mPIJ74N38htoIWwjRgGkNBga99JK7wxBCCHGGMaATjSInkPbPXDw6HaWgUIdqOVPJoSgk/vxVjSRBit11xV2E+IQw8aeJbPh3A3+c/IMCawEv/PICd3a7E2+DN408GwGQlp9GblEu/9/encfXeKb/A/88Z8lJcrLvyEZKEEHVVqlQsZXWWpQxLaraDroYpkO/Q4vSMbUU3W2dUvPL6DDVsbfKDEpotbETgoREIpKTcyLJWe7fH2lORbYjZ0nOyef9euX1iue5nifXuXqa5L5y3/ez8aeNuFZwDUsOLsHCfgsR4BGAZr7OOytZIQNCfGW4mmOscFwCEKCWQSH7LS7AS4bCu0aIGuIcGWPLnIjshU0QF1a+H4jOt6yx4auWoKiiq1q+FKZndM8a73f31jFor24HICH4kXmQJHZoiYiIiFydV9RQ5P68DJLSBM+uFfcEEcafodfNsulS6AEtB6DUWDb74a6h7Ok01wquYdH3i2q87lT2KQzbOAwAkDYzzWb5OJokSRjWzQPvf6OtcFwA8PaUYdPBIvMxb08ZBIw1xjkyxtp7De3mwTEG2R2bIC6sfCZIkbps2mKgV+UNT4UQFjVBhMmInBPzAQA+MWPgHhBn63SJiIiIqAEy6TWoblwqyU0wltyx+X5wywYtw4wdMyodj/SLRIBHAAAg724eruVfq/JaZxcXoUR0sBzp980GSb2qt+h6S+IcGVNbnCQBUUFyxEVwKSzZH5sgLqx8JojWvewHRVWboqblpeFm4U24yd3QJbz6qYyaK1tQcuc0ZEpvBHb4o30SJiIiIiKnU3j6NNx7trPpPYe0GYL1J9YjNTvVfCw+NB5bx281zxQQQmD4xuEVYgAg7XYahBBOPaNAkiQMvW82SLdWbgjxrfxHzVsFRhy9UFpjnCNj6nIvITgLhByHTRAXZm6CqPwBUfWmqOWzQLqGd4WH0qPq+5RqcPvnsp3AA9q9AoV7kJ0yJiIiIiJnk5mcDKVHDPw6d7bZPfWmslkD8aHxGN1+NJJ/STYfL3+6TIWY+NFYdWQVbulu4YOjH+Cm9iYW9V8Epdx5ZxbcOxskOliO55PUVTYJhBDIvmOsMc6RMQ9yr9SrehSVCPipJc4CIYfhtjMurHw5jEbuAwAIrGImiCVLYfJOrYKx+DaUPi3g1+pZO2RKRERERE7LZMLl1auR/+OPNrulm9wNyeOSsXX8VozrMA5bx29F8rjkCo/XrRDTcRwOv3QYb/d9G3JJjn+d/hde2PoCtKXaGr5KwyZJEoZ390QTfxmGd/esdpaEJXGOjHmQe3VtWfbf018t4ywQchhJCFH52UQEANBoNPD19UVBQQF8fHzqO50Hdm3dOuTu349/Jy7ADb0Ppg3yQofo335wlBhK0Gl1JxQbirHjuR2IDY4FAOh1mTCW3Pn18wxk/XcaACNCur8H3xYj6+OlEBEREVE9Kc47heu7nqr2vHQ3AZrvJUhyOVq8+ip8H37YgdlVtj9tP6Zvn467hruIC4nDmhFrEOIVUq85UdWuZBuw6CsNvNwlLJvox0YIWcXS8TuXw7iw8pkgBaayZS73L4c5nnkcxYZihHqFolVQKwBlDZCr25MgTCWV7pdz7E14hna3+cZXRERERNRwyVX+kGSqKn8/BADhcQg+PftD898ipK1YAbeAAOgLCqAKC0OT4cNt+ghdSzwe8zi+HPMlnv/X8zh96zSe/vJpTH90OtakrMHcpLlIiEpwaD5UvYggOZRyQFsskF1gQphf5f1FiGyNy2FcmFGng0GmhM5Ytr7u/iZI+VKYx6IeM3ddjSV3qv8BZyoxzxAhIiIiosZBqW6GqKe+RcTA7RU+wgf8G95RwwAAwmsP1J30gMmE0txcCL0exRkZuLJyJe6kpDg85/ZN2mPL77Yg3CccmZpMzNkzB5fyLmHJwSVIzUpFZkGmw3OiyhRyCVHBZX+Xv5xlqOdsqLHgTBAXZtDpoHXzAwColICnquL0MnMTJPoxR6dGRFQlIQSKcnMBAJ5BQZwWS0TUQCjVzaqcDezeYxnk7v7IP78estCjcGvdFIYb9yw9kSRk/WeTw2eDAECUXxQyNBkAAJMwAQBOZZ/CsI3DAABpM9McnhNV1iJMgUtZBlzONqBHa1V9p0ONAJsgLsyo00Gr+vXxuF7yCoOJW9pbOJdzDhIkPBbFJggRNQz6oiK8F1L2y/NsrRZuanU9Z0RERDWRJAlBnf4CYdKj4OJGqJrfgKr5jQoxwngSJfmvQeUX7fD8lg1ahhk7ZlR5nBqGFqFlQ9I0zgQhB+FyGBdm0GpR6OYPoPKTYdadWAcAiPaLRoBngMNzIyIiIiLXIEkSvFuMrv68XODSkrdw+9AhCJPJgZkBQ9oMQXxofKXjp7JPmWeHUP2KCStrgmTmGVFcymd2kP1xJoiLEgYDTMXF0AaVNUHK9wPJLMhE3t08JKeWPWu9sLQQqVmpCPAIQDNfbnhKRERERA+utuWL+oICXP34Y9zauRO+nToh//hxlGRl2X3zVL1JDwCID43H6PjR+OCHD5ClzcK6E+twS3cLSwYugUrBJRj1yU8tQ4CXDHlaE9JzDGjdTFnfKZGLYxPERRl0OgCAVlWxCZL4WWKFuNyi3ArrImVufgAkAJW7sJJMBfmv9yMiIiIispT/4xG4s+cuirPPo+Tb0wAAyQMo1eTj6oZFMOqnIKjHEzb/um5yNySPS4ZSpoQkSRjbYSy+Ov0V/m/P/+Gbc98gR5uDj4d9DB/36h+nSfbXIlSBPG0p0rLYBCH7YxPERRnLmyAeQQCAgF+Xw9S2LrK04DwAAcjd0SxxDWQqX3OMXOXPx+MSERER0QMrLd0O38FR0BdegySr/Me2vMvT4NvhoF1+13STu5k/lyQJT7d7Gk28m+AP//4DjmYcxejNo7Fu5Do09Wlq869NlokJU+B4WimfEEMOwT1BXFR1M0GGtBmCYHVwhdj40HgMaTMEQgjkpa4EAPi3ehaeTRLgHtDO/MEGCBERERHVhaTwhEF3tcoGCABIMhP0BTcdlk9CVAL+39j/h1CvUFy8fRFPf/k0Nv+8GQPWDcChq4cclgeVafHrviCXsw0QgvuCkH2xCeKijDodBIBCZdlMjvKNUfUmPTTFGgDAqHajzBtF6U16FN08gJK8XyDJ3eHX5oV6yZuIiIiInI9c5Q9JVvXeGpJMhYj+/4JPzDM13iNt+XLc2rMHJoNjZgO0Dm6NLeO2INovGtnabMzdNxeX8i5hycElSM1KRWZBpkPyICAiSA6FDNAWC+RouGEt2ReXw7gog1aLuwovGCUFJJRtOAQANzQ3UGIsgVySY3bv2fBR+UBv0kMpUyI79X0AgG/L30HhHlSP2RMRERGRM1GqmyHqqW9hLLlT6Vz5kmrflr+DJu0f1d7DeLcQGV98gZzdu9F09GhAknBz27YaN1C9k5KCm1u31nmT1aY+TZGenw4A5qfFnMo+VWHPPLI/pVxCVLACadkGpGUZEOIrr++UyIWxCeKijDqdeSmMr1qCQl62Y/d3ad8BALpGdIWve9ksETe5G3Q3D6L49klIchX827xYP0kTERERkdNSqptZtXxanXAZ+vRwFF8wIH3Ne5DcymaEVLeBau7hncjY9CkgSZA8RJ03Wa1tzzxyjBZhZU2Qy1kGPBrLJ/aQ/bAJ4qIMWq25CRLo9VsndV/aPgBAUkyS+di9e4H4PvQ7KDwq7hlCRERERGR3Qgdl1Hkoo9wgTKWQqli4n5c2Fdc/L5vpoX4sBeoE6zdZHdJmCNafWI/U7NQKx6/lX4MQotbH/5JtxIQpsPfnsn1BiOyJe4K4KKNOh0JVAIDfngxTUFyA4xnHAVRsgtzNPozi3BOQZG7wbzPF8ckSEf1KoVLh6eRkPJ2cDIWKfwUiInIlNe8b4oagTn+Bm18sgKobIAAgyQWUEZlQRmRCkle/yaqhOM/ivPQmPYCyhwUs6LsAIeoQAMCKwyvwzvfvmJfJkH01Dy37+3zGbSNK9NwcleyHM0FclEGng9YtFMBvT4Y5cOUAjMKIVkGtEOkXCaBsFsjtX/cC8XloLBSeofWTMBERAJlCgbhRo+o7DSIisgNL9g3xi52I/AufI/fE29XeR9Wi9g1L0z/6CE0GTYbvww/XOpPDTe6G5HHJUMqUkCQJYzuMxboT67Do+0VYf2I98u/m492B70Ih49DJngK8ZPBXy3BHZ0L6LQNimynrOyVyUfw/2UUZtVoUqloD+G0myL5LlZfC3L11FMU5KZBkbgho+5LjEyUiIiKiRqO2fUMkSYJHcOca7+ER1hMAcDfrv9XG6HUXcHn5cni2DIG6dTgKz52H/vZtKAMDEfz44/Dr9FiFPNzkbhVyeL7z8wjwCMAbu97A1jNbUVhSiJVPrYRKwVmK9tQiTIETaaW4nM0mCNkPmyAuyqDTQetZtidIgJcMpcZSHLhyAADQ96G+5ri88lkgMaOh8AxzfKJERERERA8gqOOfAADXd1XfBPGIT4MxIgcyn0IUGwSUDwHKh8rO5V/fjfwMJaKH7K+xITM8bjh83H0w7etp2Je2DxO/mohPhn0Cb5W3TV8P/aZFqBwn0oC0LO4LQvbDPUFclFGng9bt1yaItwwpGSnQlmoRq/ZHrJsMxXmnUHDpH7h76wdAUsA7eng9Z0xEREREZCOSAnI/DSRZNXtLCH2Vy3LulxSThA1Pb4CXmxeOXj+K8cnjsfP8TgxYNwCHrh6q9rpDVw/ZJKaxiQkr+xv95WwDhOC+IGQfnAnioop1xbjrVtalDvSSYf3P+xAsF1jqewcZu4dWDBYGZH47DlFPfWvVY82IiIiIiKxVvoGqMJVUOifJVJD/+gTEmmKaJW3CnbNroMvYVe3X0aWlQeUfV+ueId0iumHTmE14NvlZnMo+hRk7ZqDUWIo5u+fg6finoXZTI8Cj7IEEeXfzoCvVYUvqFmRoMiyKWXJwCRb2W4gAjwA0823cv4tHBiugkAGFdwVyNSYE+8prv4joAbEJ4oKEyYQCQ9kaOpUC8HADvk37Fr4yQIGqd7cWphIYS+6wCUJERERE9cqSDVQB1BojyVU1NkGub1iP27tT0WTkSBi0WtzcuhUlWVlQhYWhyfDh8O/SxRzbLrQdCkoKAAClxlIAQIYmAysOrajxtVgScyr7FIZtHAYASJuZVmOsq1PKJUQEy3El24i0bAObIGQXbIK4IFNxMbTK8v1AJFy4fQGZmky0cXcDULlbTkRERETUkNS2gaqlMTXx6HQOJWk6XFpxEpLCCACQPIBSTT6ublgEo34Kgno8YY5fNmgZZuyYUek+rQJbIcSr7LG6t7S3cOH2hTrFvPfEe3V+La4kJlSBK9lGXM4yoHsrbkRLtscmiAsyaLXQ/jpNMNBHYX4qTMcmHYGSo/WYGRERERFRwyDzKIF7u0sQAqhqRUze5Wnw7XDQ3GgZ0mYI1p9Yj9TsVHNMfGg8to7fal5SI4TA8I3DHzgGAPZe3ItBsYMa/RNoWoQpgF9KkJbNzVHJPrgxqgsy6nQoVP32ZJhv074FAHQN71qfaREREREROUz53iJVkWRu8G/3CuTuQVU2QMpiTBWW2+hNegBlTY0F/RYgPjS+wvG6xkT6RUKChN2XduOFrS9AV6qr4yt2DTGhZX+nz8g1okTPzVHJ9iTBbXerpdFo4Ovri4KCAvj4+NR3OhbTpKbis6+u41xId/TtaMScw90BAP8buwHaA89We13EwO1wD2jnqDSJiCoxGY249t+yRx5G9uwJmZxrgYmIqO70uswa9w25m3sSGXuqf0qiyvQymg59GQrvsgcOlBpLoZQpIUkShBDQm/Rwk7tVuKYuMQfTD2La19NQpC9CxyYdsWbEGvh7+NugAs5HCIE//T0f+TqBWcO80aqpsr5TIidh6fidy2FckOGex+PeKDoDoGwpjL9HALTVXHPvTttERPXFUFyMzx9/HAAwW6uFm1pdzxkREZEzq23fEElW83BIe2MLTs8+i9CBI+H/WDtoUo8iZ/9+6G/fhjIwEMGPPw6/To+Zv4Zel4mCH//3wDFtH38cG4csw6T//Bknb57E2H+MxYanNyDMO8x2xXASkiShRagCP17WIy3LwCYI2RyXw7igsuUwfgCAM7llzx1PiklCwaVNAACldwuE99+GiIHbzR98PC4RERERUUXK8Bx4dD2I3NR3cPWbJORf/zOUD+2GZ7fjUD60G/nX/4z0rx+HXpcJvS4T6V8/XucYzx+mYvPQ9xHqFYqLty9i9ObR2HJqCwasG4BDVw/VmOehq4dqjXNkjLX3igkra05d5r4gZAdsgrggg1Zn3hj1eFbZfiBJIdHQXPoHACC027vwCOoA94B25g82QIiIiIiIKlJ6x0BSmKCKyoIkM1UdJPS4sXUTbmzdBAi9VTGRaj8kj01GM59myNRkYs7uObiUdwlLDi5BalYqMgsyK1ySWZCJ1KxULDm4pNo4R8bY6l4tft0X5HKWAdy9gWyNe4LUwFn3BDm/cQve0/QBILC/qAfCfcPwRXQYinOOwSvqKTRJWFnfKRIRValUp8NiLy8AXA5DRET2p9dl4ur2JAhTSaVzkkyFyCf3oTT/HHJ+XASD9kq19zEWegAA5N53rYop36Mv5r2YamM6hHUwf/5z1s+1xjkyxhb3SpuZBr1BYPqaOzCagEXjfRHswz3CqHbcE6QRy9OV9bUkFEDAgAnNYlCcsx+S3B1BHd+o5+yIiIiIiBoGpboZop76tsbNU928wiH3CEXG7iHV3qemxsaDxGhOnoSqd1ssG7QMM3bMqDKmpgbCg8Y5MsaSuGWDlgEAlAoJkUFyXLllxOVsA5sgZFNsgrigO8UyQAloTZlQSQLd7/4CAPBv+xKXvRARERER3aO2zVOBss06axL08P8BAHJ/WmhVzO2zf8XtIwfR84lnMDA0BpfyLpnPhXk1wYTur8DkHmQ+JivOxYYfViJLe7PaOEfGWHMvT4UafZr+9qTKFqGKsiZIlgHdWlb9qGOiuuBymBo463KYL//2NfarH0O2YS+6es7DGM8SKDybIurJfZApPOo7PSKianE5DBERNUTFeadwfddT1Z6PGLgdAKyOKScEUGXfReaG6Ke+g1LdrGyT1e19AFNptXEAHBZji5z0kBA+aA98/B7CsYsl+GyvDlHBcvzfKN8qikFUEZfDNGIFJncAgAw3McKzbEfloIdnswFCRERERFQHcpU/JJmq2r1D5L8+lMCaGMjc4BmWhKLMXZCkav5ObSpF1o7PIZM1gcl0s+pmwz1x5Z87IsYWOSkhMG/n61gwcjNiwsrGLhm3jSg1CLgpap6NQ2SpBtkE2bx5M9577z2cOXMGarUa/fr1w7vvvouoqCiLrj9x4gTefPNNHDlyBCaTCY888gjmz5+PxMREO2feMBQITwDAI6qbUMII9+Au8IocXM9ZERERERE5J0v2DgFgkxhtxj7cPPhCtbkU6z+zKGdL4hwZY2ncmVunMeGfE7B2xFr4ekooKBK4mmNAyyZKi74GUW0aXBNk9erVmD59OhISErB8+XLk5uZixYoVOHjwIFJSUtC0adMar09JSUGvXr0QEhKCv/zlL1CpVPj000+RlJSEnTt3om/fvg56JY6XfesGNJpC5KjKpv7EKBXIMrRG8+av1LqOkYiIiIiIqmfJ3iG2iFF4htWciJAASAAEUN2MEXMcHBhjm5zCFe74/uZPmJ78Mlp7/hUFRWocPXUD0BUDAHx9vBASUla/8vHP/e6NsTTOkTHOnJMraFB7gty+fRvR0dFo1aoVjh49CoWirEdz/PhxdO3aFZMmTcKaNWtqvMejjz6K1NRUnDlzBpGRkQCAgoICxMXFwdPTE+fPn7e4IeBMe4IUaTIxe1MRikRgpXNq6TYW/c4Tnj6u88YlItckTCbknD0LAAhu0waSTFbPGRERETmWJfuPuAe0c9g+JZbG2ConIYBjd1VI0W+vcWwDwKLxjyXjJEvuZasYZ86poY8nLR2/N6jfLv/9739Dq9XilVdeMTdAAKBz585ITExEcnIySkurWWMG4PLly/jhhx8watQocwMEAHx9fTF58mRcvHgRR48etetrqC+S/g58ZFkATPedMcFblgVJX3nKHRFRQyPJZAiJi0NIXBwbIERERI2QJAFdPUpqHdtYOv6xJM6RMc6ck6toUMthjh07BgDo0aNHpXM9evTAgQMHcO7cObRv375O15fHdO/evcrrS0pKUFLy2yZFGo3mwV5APZIkoKf7R/inbvV9Z2To6f4RJGlOveRFRERERESWs3QTVkds1vqgMbbIKbjLcuQeW1Ht2OauyQ/zknUAABP8UPnv+r/FCByFZEGcJfeyVYwz5uRq48kG1QTJzMwEAISHh1c6V34sIyOj2iaIpddXZ/HixXj77bcfLOkGJFpxBP6ydNwxRQMAJBgRKj+HaMWR+k2MiIiIiIgsYukmrI7crNXSGFvlpPKPgHHnUwiTn0aWsS3K9hspUyBqX5JhSYwt7+WqObnqeLJBNUGKiooAACqVqtI5d3f3CjH2uH727NmYMWOG+d8ajQYREREWZN4wSBLQy30VthUtBQAIyH/t2tVzYkREREREZDFLNli1NM6RMba8V1Uz3RNVKxGmOIsC9aMAAF/dEWQZ2uBgyStVxhjlPpAbNbXGWXIvW8U4U06uOp5sUE0QT8+yjVtKSkrg4eFR4dzdu3crxNR2/f0suV6lUlXZQHEmLZXf/9oxjUOY/LTLde2IiIiIiKhxiFYcqTC26eb+OSQJiOjzFwDA9V3vI0pxDBcMSVXG/LZZa81xltzLVjHOlpMrjicb1K5zzZqVdQOrWrJS01IXW13vzMrX30kSkOi+GoGyy0h0Xw1Jqrj+joiIiIiIqCGzZGxj6fjHVvdiTq4znmxQj8hdu3YtJk+ejA0bNuC5556rcK537944ceIEbt++DTc3tyqvT0tLw0MPPYQJEyZg/fr1Fc699dZbePvtt3H48GE8+uijFuXjTI/IBQC9LrPW9XdEREREREQNnSVjG0vHP7a6F3Nq2CwdvzeoJkhubi6ioqLQunVrHD161PyY3OPHj6Nr166YOHEi1q5dCwC4efMmCgoKEBkZWWGJS7du3XD69GmcPXvWvJ+HRqNBXFwcVCoVLl68CMnCRU3O1gQhIiIiIiIiaowsHb83qOUwQUFBWLRoEX788Uf07t0bn3zyCd555x0MHDgQoaGhmD9/vjl29uzZaNOmjfmxuOVWrlwJo9GInj17YtmyZVi9ejUSEhJw8+ZNfPjhhxY3QIiIiIiIiIjItTSojVEB4NVXX0VQUBCWLl2K1157DZ6enujXrx8WL15s3vOjJt26dcPBgwfx5ptv4q233oLRaETnzp2xb98+9O7d2/4vgIiI6syo1+PEp58CAB6ZMgVypbKeMyIiIiIiV9KglsM0NFwOQ0TkWKU6HRZ7eQEAZmu1cFOr6zkjIiIiInIGTrkchoiIiIiIiIjIXtgEISIiIiIiIqJGgU0QIiIiIiIiImoU2AQhIiIiIiIiokaBTRAiIiIiIiIiahTYBCEiIiIiIiKiRoFNECIiIiIiIiJqFNgEISIiIiIiIqJGQVHfCTRkQggAgEajqedMiIgah1KdDsW/fq7RaOBmNNZrPkRERETkHMrH7eXj+OpIoraIRiwjIwMRERH1nQYRERERERERWeD69esIDw+v9jybIDUwmUy4ceMGvL29IUlSfadjMY1Gg4iICFy/fh0+Pj71nY7LY70djzV3LNbb8Vhzx2K9HY81dyzW2/FYc8divR2vIdZcCIHCwkI0bdoUMln1O39wOUwNZDJZjR2khs7Hx6fBvCEbA9bb8Vhzx2K9HY81dyzW2/FYc8divR2PNXcs1tvxGlrNfX19a43hxqhERERERERE1CiwCUJEREREREREjQKbIC5IpVJh3rx5UKlU9Z1Ko8B6Ox5r7list+Ox5o7Fejsea+5YrLfjseaOxXo7njPXnBujEhEREREREVGjwJkgRERERERERNQosAlCRERERERERI0CmyBERERERERE1CiwCUJEREREREREjQKbIE5k8+bNeOSRR+Dh4YGgoCCMHTsWV69etfj6EydOYODAgfD19YW3tzd69+6NgwcP2jFj52ZNvZOTkzFx4kS0b98eCoUCkiQhPT3dvgm7gLrW/M6dO3j//ffRv39/REREwMPDA7GxsZgyZQquX7/ugMydU13rrdfr8dJLL+GRRx5BUFAQVCoVmjdvjjFjxuDkyZP2T9yJWft9/F6jR4+GJElo3bq1jbN0HdbUu3fv3pAkqcqPbdu22TdxJ2bte9xoNOLjjz9Gt27d4O3tDS8vL8THx2PBggV2zNp51bXe33//fbXv7/KPQ4cOOeAVOB9r3uNCCHzxxRd49NFHERgYCB8fH8THx2PRokXQarV2ztw5WVNvvV6PRYsWoU2bNlCpVAgMDMTIkSNx7tw5O2ftvBYvXoxRo0ahRYsWkCQJ0dHRdbpPQx938ukwTmL16tWYPn06EhISMH78eOTm5mLFihVQqVRISUlB06ZNa7w+JSUFvXr1QkhICKZNmwaVSoVPP/0U586dw86dO9G3b18HvRLnYG29e/fujaNHj6JDhw7Iz8/H+fPnceXKlTp/I2kMrKn5rl27MHjwYPTp0wdJSUkICgrC6dOn8cknn8DNzQ2HDx9G27ZtHfhqGj5r6q3T6dCrVy8kJCSgefPm8Pb2xrVr17B+/XpkZWVh586dSEpKcuCrcQ7Wfl+513/+8x8MGTIEKpUKkZGR/IWuCrb4Pn769GksX768ynPh4eH2St1pWVtzvV6PESNGYNeuXXjmmWeQkJAAmUyG9PR05OTk4LPPPnPQK3EO1tQ7Ozsbe/furXS8pKQEU6ZMQVBQEDIyMqBUKu35EpyOte/x2bNn491330WfPn0wfPhwyOVy7N27F1u3bkViYiIOHDjgoFfiHKyptxACTz75JHbs2IGhQ4eif//+yMnJwYcffoiSkhL+blgNSZIQEBCATp064cSJE/Dx8XngP+Q6xbhTUIOXm5srvLy8RKdOnYRerzcfT0lJEZIkieeff77We3Tv3l2o1Wpx9epV87H8/HzRrFkz0bJlS2EymeySuzOyRb2vXr1qvnbq1KkCgLhy5Yq9UnZ61tb8ypUr4uLFi5WO7927VwAQTz/9tM1zdma2eI9XJTMzU8jlctGvXz9bpeoybFnzwsJCERkZKaZNmyaioqJEbGysPVJ2araod69evURUVJQds3Qttqj53LlzhUwmE7t27bJnqi7BXt/Hv/zySwFAzJw501apugxra67X64Wnp6fo1KmTMBqNFc4NHTpUABBnz561S+7OyNp6b9u2TQAQU6ZMqXA8LS1NeHh4iKSkJLvk7ezS0tLMn8fFxdXp56AzjDvZBHECa9euFQDEhg0bKp3r1auX8Pb2FiUlJdVen5aWJgCICRMmVDo3b948AUAcOXLEpjk7M2vrfT82QWpn65rfKyAggIPE+9ir3kajUXh7e4vOnTvbIk2XYsuav/rqq6JJkyaioKCATZBq2KLe5U0Qo9EoCgoKKg1aqCJra67VaoWPj48YOnSoEEIIk8kkNBqNvdJ1evb6Pt6nTx8BQJw7d84WaboUa2teVFQkJEkSgwYNqnTu5ZdfFgAqDBobO2vr/eqrr1Y7xunfvz/rbYG6NEGcZdzJPUGcwLFjxwAAPXr0qHSuR48eKCwsrHEqdG3X3xtD1tebHpy9al5QUIDCwkKEhIRYnaMrsVW9jUYjcnNzkZ2djePHj2P8+PEoLCzE4MGDbZ6zs7NVzVNSUrBq1SosX74cPj4+Ns/TVdiq3pmZmfDy8oKvry/UajUGDRqE48eP2zxfV2Btzf/3v/9Bo9Gga9eumDlzJvz8/ODj44OAgABMnz4dRUVFdsvdGdnj5+aVK1ewf/9+PPbYY4iNjbVJnq7E2pp7eHigR48e2LVrF5YsWYJLly4hPT0dn332GdavX4/JkycjMjLSbvk7G2vrXVxcDADw9PSsdK78GMc/tucs4042QZxAZmYmAFS5/rj8WEZGht2ub2xYL8ezV80XLlwIvV6P5557zroEXYyt6n327FkEBwcjLCwMXbp0wTfffIM//elPePPNN22bsAuwRc0NBgNeeOEF9O3bF2PGjLF9ki7EFvWOjo7GrFmzsHbtWnz11VeYOXMmDh06hISEBHz33Xe2T9rJWVvz8sHMihUrsGnTJixcuBBbtmzB4MGDsXr1agwZMgSC29iZ2ePn5rp16yCEwOTJk61P0AXZouZffvklEhMT8cYbb6Bly5Zo3rw5XnrpJcyZM4d73tzH2nqX7/dx//froqIiHD16FABw7do1m+RKv3GWcZSivhOg2pX/9UOlUlU65+7uXiHGHtc3NqyX49mj5snJyVi6dCn69euHiRMnWp+kC7FVvZs3b469e/eitLQUly5dwpdffgmdTofS0lJupncfW9R86dKlOH/+PL766ivbJ+hibFHvDRs2VPj3iBEjMH78eHTq1AkvvfQSLly4YJtkXYS1NS8sLAQA5OXl4ZdffjEPYEaOHAkA2LhxI/bs2YMBAwbYNG9nZeufm0ajERs2bICPjw9GjRplmyRdjC1q7unpidjYWERGRmLgwIGQyWTYtm0b5s6dC6PRiLfeesvmeTsra+s9fvx4LFy4EHPnzoVarUbfvn2Rm5uLefPmITc3t9brqW6cZRzFmSBOoHzKVklJSaVzd+/erRBjj+sbG9bL8Wxd8x07duD3v/89Hn74Yfzzn/+ETMZvdfeyVb3Lf6kYNGgQXnnlFezbtw979uzBiBEjbJuwC7C25mlpaXj77bcxZ84cxMTE2CdJF2Kv7+OxsbEYPXo0Ll68iIsXL1qXpIuxtuYeHh4AgG7dulV6YsOkSZMAAPv377dJrq7A1u/x3bt3IyMjA2PHjuXvONWwtuZFRUXo0aMHCgoK8Pnnn2Ps2LEYM2YMNm/ejIkTJ2L+/Pl8zPw9rK13QEAA9u7di+bNm2PKlClo0aIFunbtCo1GgzfeeAMAuKzUDpxlHMWRgRNo1qwZgKqnDtU05chW1zc2rJfj2bLmu3btwogRI9C6dWvs2bMHvr6+tkvURdjrPe7l5YURI0Zgz549SEtLsy5JF2Ntzf/4xz/C398fY8aMQXp6uvnDYDBAr9cjPT0d2dnZ9kneCdnz+3j5o85zcnLqlpyLsrbm5eeaNGlS6Vz5sby8PKvzdBW2fo+vXbsWALgUpgbW1nzLli24ePFilTNtxowZAyEEH5F7D1u8xzt06ICff/4ZFy5cwIEDB3DhwgX873//Mw/QW7dubeOsyVnGUWyCOIEuXboAAA4fPlzp3OHDh+Hl5VXj/8S1XX9vDFlfb3pwtqr57t27MXz4cLRq1QrffvstAgMDbZ6rK7Dne7y8y8/BSkXW1jw9PR03btxAbGwsmjdvbv7IzMzE5cuX0bx5c+59cw97vsfLZ4CEhYXVPUEXZG3Nu3XrBgC4fv16pXPl6/ZDQ0NtkapLsOV7/NatW9i+fTvat2+Pzp072zRPV2JtzcsHgHq9vtK58mMGg8EWqboEW77HW7ZsicTERLRs2RIAsHPnTvj4+CAhIcF2CRMAJxp31vPTacgCOTk55ueKV/Wc7EmTJpmP3bhxQ5w9e1bodLoK9+jatatQq9Xi2rVr5mMFBQUiPDxcxMTENIjnNTcUtqj3vfiI3NrZoua7d+8W7u7uIj4+XuTk5Dgsd2dkbb1v3bpV5eNCb968KZo0aSK8vLxq/H+iMbK25t99953YunVrpY/g4GDRrFkzsXXrVnH48GGHvqaGzNp65+XlVfnoxZSUFKFUKkVcXJx9X4ATssX38cTERCFJUoXHJ5pMJjFkyBABgO/xe9jyd5W//e1vAoBYuXKl3fN2ZtbWfNu2bQKAeOKJJyrde/jw4QKAOHjwoH1fhBOx9e/j5VauXCkAiHnz5tkjbZdS2yNynXncySaIk1ixYoUAIBISEsTHH38sFi5cKAIDA0VYWJjIyMgwxz333HMCgNi/f3+F63/44Qfh7u4uoqKixNKlS8WqVatEu3bthFwuF7t373bwq2n4rK33gQMHxIIFC8SCBQtEt27dBADxxz/+0XwsPz/fwa+o4bOm5ikpKcLd3V2oVCqxfPly8cUXX1T6oIqsqffy5ctFVFSUeO2118T7778vPvroI/H666+LgIAAIUmSWLt2bT28oobP2u8rVYmKihKxsbF2zNp5WVPvrVu3ipCQEDF16lSxfPly8eGHH4opU6YINzc34enpycF4Nax9j//888/C29tb+Pj4iDfffFOsXr1a9O/fXwCoMOChMrb6ntKmTRvh7u4u8vLyHJS587Km5gaDQXTt2lUAED179hTLly8XK1asEL179xYAxJNPPlkPr6hhs/Y9/sQTT4ipU6eKDz74QHz44Ydi2LBhAoAYPHiwKC0tdfCrcQ5///vfzeOVkJAQ4efnZ/73qlWrKsQ687iTTRAnsnHjRvHwww8Ld3d3ERAQIMaMGSMuX75cIaamH3THjh0T/fr1E97e3sLT01MkJiZa9Et2Y2VNvefNmycAVPvBWSFVq2vN169fX2O9OemtanWt9/Hjx8W4ceNETEyMUKvVQqlUivDwcDF69Ghx6NAhB78K52Lt9/H7sQlSs7rW+8yZM2LUqFEiJiZGeHl5CaVSKaKiosSkSZPEhQsXHPwqnIu17/FTp06J4cOHC39/f+Hm5ibatm0rli9fXuXsM7K+3ocOHRIAxLhx4xyUsfOzpuY6nU789a9/FR07dhS+vr5CpVKJuLg48c4771Q5+4ysq/f8+fNFXFycUKvVQq1Wi86dO4sPPvhAGAwGB74C59KrV69qf5e+f1aIM487JSH40HUiIiIiIiIicn3cGJWIiIiIiIiIGgU2QYiIiIiIiIioUWAThIiIiIiIiIgaBTZBiIiIiIiIiKhRYBOEiIiIiIiIiBoFNkGIiIiIiIiIqFFgE4SIiIiIiIiIGgU2QYiIiIiIiIioUWAThIiIiIiIiIgaBTZBiIiIyKG+//57SJKEDRs21HcqldgztwkTJkCSJIvjJUnChAkTbJ4HERFRY6ao7wSIiIjIuT3IwP7KlSt2zISIiIioZmyCEBERkVW++OKLCv8+e/YsFi1ahOHDh2PEiBEVzgUHByM9Pd2B2RERERH9hk0QIiIissr48eMr/Pv777/HokWL0L59+0rnbEWr1cLLy8su9yYiIiLXxT1BiIiIqN6sWbMGbdu2hUqlQlRUFJYsWVIpJjo6Gr1798ZPP/2EAQMGwNfXF/Hx8ebzFy9exO9//3s0adIEbm5uiI6OxqxZs6DT6Src5/r163j++ecRFRUFlUqFwMBAdOnSBZ999lmdcwOA7du3o2fPnvD29oZarUbXrl2xefNmi2tw5MgRJCYmwsPDA0FBQXj22WeRk5Nj8fVERERkOc4EISIionrx0Ucf4datW5g8eTJ8fX2xceNGvPHGGwgPD8e4ceMqxF67dg1JSUkYNWoURo4cCa1WCwA4ceIE+vTpAz8/P7z44oto1qwZfvnlF6xcuRKHDh3CgQMHoFQqYTAY0K9fP2RmZuLll19GbGwsNBoNTp06hYMHD+KFF16oU26ffvopXnzxRbRs2RKzZ8+Gm5sbNm7ciHHjxuHKlSuYM2dOjTU4evQo+vTpAw8PD8ycORMhISHYtm0bBg4caKMqExER0b3YBCEiIqJ6cf36dZw5cwZ+fn4AgEmTJiEqKgqrVq2q1AS5cuUK1q1bh4kTJ1Y4PmnSJISFheH48ePw9vY2H+/Tpw9GjBiBTZs2YcKECThz5gzOnz+PJUuWYNasWTbJLT8/HzNmzEB0dDRSUlLg6+sLAPjDH/6ARx99FPPmzcP48eMRGRlZ7dd5/fXXodfrkZKSgnbt2gEApk6dimHDhuHHH3+sNU8iIiJ6MFwOQ0RERPVi4sSJ5iYDAHh6eqJ79+64ePFipdjAwEA899xzFY6lpqbil19+wTPPPIOSkhLk5uaaPx577DGo1Wrs2bMHAMwNiu+++w7Z2dk2yW3v3r3Q6XSYPn26+f7lsTNnzoTBYMDXX39d7de4desWjhw5gieffNLcAAEAmUyGP//5z7XmSERERA+OTRAiIiKqFy1atKh0LDAwELdv364yViar+GvL2bNnAQDz589HcHBwhY+QkBDodDpzwyMqKgpz587Fnj170LRpU3Tq1AmzZs3CDz/8UOfcLl++DACIi4urFFu+Z0l5TFXKz7Vp06bSubZt21Z7HREREdUdl8MQERFRvZDL5RbHenp6VjomhAAAvPbaaxg8eHCV1/n7+5s/f/vttzFhwgTs2LED//3vf7F+/Xq89957mD59OlauXPnAuZV//Qc9dz9Jkiw6RkRERNZjE4SIiIicUqtWrQCULR/p27evRdc0b94cU6dOxdSpU1FSUoKhQ4di1apVeP3119G8efMH+voxMTEAgNOnT2PAgAEVzp0+fbpCTE3XnzlzptK58uuJiIjItrgchoiIiJxSx44dER8fj08//RSXLl2qdN5gMCAvLw8AUFBQAL1eX+G8SqUyL2Upj3sQ/fr1g1qtxurVq6HRaMzHi4uLsXTpUigUCjz11FPVXh8cHIwePXrgm2++walTp8zHTSYT3n333QfOh4iIiGrHmSBERETklCRJwt///nf06dMHHTt2xKRJkxAXF4eioiJcunQJ//rXv7B48WJMmDAB+/fvx5QpUzBy5Ei0atUK3t7eOHnyJD755BO0b98eHTt2fOCv7+fnh6VLl+Kll15Cly5dMHHiRCiVSmzcuBEnT57EO++8U+OTYQBg2bJl6N27NxITEzFt2jQEBwdj27ZtyM/Pr1tRiIiIqEZsghAREZHT6tixI3766ScsXrwYX3/9NT7++GN4e3sjOjoaEyZMQFJSEgCgQ4cOGDFiBA4cOIBNmzbBaDQiIiICM2fOxKxZsx5of5J7vfjii2jSpAmWLFmCBQsWQAiBdu3aYdOmTZUe81uVbt264bvvvsMbb7yBv/3tb/D09MTgwYPxj3/8AyEhIXXKiYiIiKoniQfZuYuIiIiIiIiIyElxTxAiIiIiIiIiahTYBCEiIiIiIiKiRoFNECIiIiIiIiJqFNgEISIiIiIiIqJGgU0QIiIiIiIiImoU2AQhIiIiIiIiokaBTRAiIiIiIiIiahTYBCEiIiIiIiKiRoFNECIiIiIiIiJqFNgEISIiIiIiIqJGgU0QIiIiIiIiImoU2AQhIiIiIiIiokbh/wN71fWCITUcfgAAAABJRU5ErkJggg==",
      "text/plain": [
       "<Figure size 1300x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(13,5))\n",
    "pyplot.plot(thresholds_cust, np.array(tss_score_stsf), marker='o', markersize=4, color='indianred',  label='TSS')\n",
    "pyplot.plot(thresholds_cust, np.array(hss_score_stsf), marker='*', markersize=4, color='forestgreen', label='HSS')\n",
    "pyplot.plot(thresholds_cust, np.array(gss_score_stsf), marker='s', markersize=4, color='goldenrod', label='GSS')\n",
    "pyplot.plot(thresholds_cust, np.array(mcc_score_stsf), marker='v', markersize=4, color='cornflowerblue', label='MCC')\n",
    "plt.axvline(x = thresholds_cust[hss_ix_stsf], color=\"maroon\", linestyle=(0, (5, 5)))\n",
    "plt.text(0.3, 0.3, r'Chosen threshold ='+str(thresholds_cust[f1_ix_stsf]))\n",
    "plt.xticks(np.arange(0.0, 1.1, step=0.1))\n",
    "# axis labels\n",
    "pyplot.xlabel('Threshold')\n",
    "pyplot.ylabel('Score')\n",
    "pyplot.legend()\n",
    "# displaying the title\n",
    "plt.title(\"Supervised Time Series Forest\")\n",
    "# show the plot\n",
    "pyplot.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "9048f829-501d-4ec3-912c-f1f82590b52a",
   "metadata": {},
   "outputs": [],
   "source": [
    "#Choosing the best classification threshold of the STSF model\n",
    "best_stsf = thresholds_cust[hss_ix_stsf]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "f1fbe490-04aa-4e09-baf2-e3c568df7961",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best Threshold = 0.330\n",
      "Scores -- F1_Score: 0.975 / TSS: 0.769000 / HSS: 0.831000 / GSS: 0.711000 / MCC: 0.834697\n"
     ]
    }
   ],
   "source": [
    "# evaluate the best threshold\n",
    "f1_scorex_stsf = f1_score(y_valid, to_labels(probs_stsf, best_stsf), average='weighted')\n",
    "tss_scorex_stsf = tss(y_valid, to_labels(probs_stsf, best_stsf))\n",
    "hss_scorex_stsf = hss(y_valid, to_labels(probs_stsf, best_stsf))\n",
    "gss_scorex_stsf = gss(y_valid, to_labels(probs_stsf, best_stsf))\n",
    "mcc_scorex_stsf = matthews_corrcoef(y_valid, to_labels(probs_stsf, best_stsf))\n",
    "print('Best Threshold = %.3f' %(best_stsf))\n",
    "print('Scores -- F1_Score: %.3f / TSS: %f / HSS: %f / GSS: %f / MCC: %f' % (f1_scorex_stsf, tss_scorex_stsf, hss_scorex_stsf,\n",
    "                                                                            gss_scorex_stsf, mcc_scorex_stsf))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "888def20-c5d0-4be5-9fb0-9cd5974e3b18",
   "metadata": {},
   "source": [
    "### Validation set confusion matrix on the best threshold"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "5110e245-9601-4a87-b57c-0ea16e77a8a0",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "y_pred_class_stsf = to_labels(probs_stsf, best_stsf)\n",
    "color = 'white'\n",
    "stsf_matrix = ConfusionMatrixDisplay.from_predictions(y_valid, y_pred_class_stsf,\n",
    "                                                     cmap=plt.cm.Blues, colorbar=False)\n",
    "stsf_matrix.ax_.set(title='Supervised Time Series Forest')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d62f230e",
   "metadata": {},
   "source": [
    "## Test set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "8905d4a5-b36e-43e1-a6c7-2ba44b0e15df",
   "metadata": {},
   "outputs": [],
   "source": [
    "# predict probabilities on the test set\n",
    "probX_stsf = stsf.predict_proba(X_test)[:,1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "3182c582-5683-4326-90b1-27c2715cf470",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test scores -- F1_Score: 0.980 / TSS: 0.850 / HSS: 0.878 / GSS: 0.783 / MCC: 0.879\n"
     ]
    }
   ],
   "source": [
    "# evaluate on the best threshold\n",
    "f1_test_stsf = (f1_score(y_test, to_labels(probX_stsf, best_stsf), average='weighted'))\n",
    "tss_test_stsf = tss(y_test, to_labels(probX_stsf, best_stsf))\n",
    "hss_test_stsf = hss(y_test, to_labels(probX_stsf, best_stsf))\n",
    "gss_test_stsf = gss(y_test, to_labels(probX_stsf, best_stsf))\n",
    "mcc_test_stsf = matthews_corrcoef(y_test, to_labels(probX_stsf, best_stsf))\n",
    "print('Test scores -- F1_Score: %.3f / TSS: %.3f / HSS: %.3f / GSS: %.3f / MCC: %.3f' % (f1_test_stsf, tss_test_stsf, \n",
    "                                                                          hss_test_stsf, gss_test_stsf,\n",
    "                                                                          mcc_test_stsf))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "054616e4-0d47-4d29-a4e3-4f4aedda8446",
   "metadata": {},
   "source": [
    "### Test set confusion matrix on the best threshold"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "id": "dd3d2413-29da-40ac-866e-c82f36ce5d73",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "color = 'white'\n",
    "y_pred_test_stsf = to_labels(probX_stsf, best_stsf)\n",
    "stsf_matrix = ConfusionMatrixDisplay.from_predictions(y_test, y_pred_test_stsf,\n",
    "                                                     cmap=plt.cm.Blues, colorbar=False)\n",
    "stsf_matrix.ax_.set(title='Supervised Time Series Forest')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "eee935f2-018f-47c2-8be2-aa138a3e021d",
   "metadata": {},
   "source": [
    "# KNeighborsTimeSeriesClassifier"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b5edec25-91ef-47fe-a872-449f00d0da26",
   "metadata": {},
   "source": [
    "## Training set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "id": "93ce48f1-049e-4e12-85ef-1588bba59dbd",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>#sk-09d857c1-4687-49b3-9f1c-07fa28dd1438 {color: black;background-color: white;}#sk-09d857c1-4687-49b3-9f1c-07fa28dd1438 pre{padding: 0;}#sk-09d857c1-4687-49b3-9f1c-07fa28dd1438 div.sk-toggleable {background-color: white;}#sk-09d857c1-4687-49b3-9f1c-07fa28dd1438 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-09d857c1-4687-49b3-9f1c-07fa28dd1438 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-09d857c1-4687-49b3-9f1c-07fa28dd1438 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-09d857c1-4687-49b3-9f1c-07fa28dd1438 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-09d857c1-4687-49b3-9f1c-07fa28dd1438 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-09d857c1-4687-49b3-9f1c-07fa28dd1438 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-09d857c1-4687-49b3-9f1c-07fa28dd1438 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-09d857c1-4687-49b3-9f1c-07fa28dd1438 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-09d857c1-4687-49b3-9f1c-07fa28dd1438 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-09d857c1-4687-49b3-9f1c-07fa28dd1438 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-09d857c1-4687-49b3-9f1c-07fa28dd1438 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-09d857c1-4687-49b3-9f1c-07fa28dd1438 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-09d857c1-4687-49b3-9f1c-07fa28dd1438 div.sk-estimator:hover {background-color: #d4ebff;}#sk-09d857c1-4687-49b3-9f1c-07fa28dd1438 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-09d857c1-4687-49b3-9f1c-07fa28dd1438 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-09d857c1-4687-49b3-9f1c-07fa28dd1438 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 2em;bottom: 0;left: 50%;}#sk-09d857c1-4687-49b3-9f1c-07fa28dd1438 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;}#sk-09d857c1-4687-49b3-9f1c-07fa28dd1438 div.sk-item {z-index: 1;}#sk-09d857c1-4687-49b3-9f1c-07fa28dd1438 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;}#sk-09d857c1-4687-49b3-9f1c-07fa28dd1438 div.sk-parallel::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 2em;bottom: 0;left: 50%;}#sk-09d857c1-4687-49b3-9f1c-07fa28dd1438 div.sk-parallel-item {display: flex;flex-direction: column;position: relative;background-color: white;}#sk-09d857c1-4687-49b3-9f1c-07fa28dd1438 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-09d857c1-4687-49b3-9f1c-07fa28dd1438 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-09d857c1-4687-49b3-9f1c-07fa28dd1438 div.sk-parallel-item:only-child::after {width: 0;}#sk-09d857c1-4687-49b3-9f1c-07fa28dd1438 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;position: relative;}#sk-09d857c1-4687-49b3-9f1c-07fa28dd1438 div.sk-label label {font-family: monospace;font-weight: bold;background-color: white;display: inline-block;line-height: 1.2em;}#sk-09d857c1-4687-49b3-9f1c-07fa28dd1438 div.sk-label-container {position: relative;z-index: 2;text-align: center;}#sk-09d857c1-4687-49b3-9f1c-07fa28dd1438 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-09d857c1-4687-49b3-9f1c-07fa28dd1438 div.sk-text-repr-fallback {display: none;}</style><div id='sk-09d857c1-4687-49b3-9f1c-07fa28dd1438' class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>KNeighborsTimeSeriesClassifier(weights=&#x27;distance&#x27;)</pre><b>Please rerun this cell to show the HTML repr or trust the notebook.</b></div><div class=\"sk-container\" hidden><div class='sk-item'><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=UUID('eb78a4bb-162b-405d-8a2c-43c583dbf872') type=\"checkbox\" checked><label for=UUID('eb78a4bb-162b-405d-8a2c-43c583dbf872') class='sk-toggleable__label sk-toggleable__label-arrow'>KNeighborsTimeSeriesClassifier</label><div class=\"sk-toggleable__content\"><pre>KNeighborsTimeSeriesClassifier(weights=&#x27;distance&#x27;)</pre></div></div></div></div></div>"
      ],
      "text/plain": [
       "KNeighborsTimeSeriesClassifier(weights='distance')"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "knnts = KNeighborsTimeSeriesClassifier(n_neighbors=1, weights= 'distance', distance=\"dtw\")\n",
    "knnts.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9d9921a2-f2b5-4d0a-92e2-4a1fced4d154",
   "metadata": {},
   "source": [
    "## Validation set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "id": "e871a45f-2945-4e77-a7c8-36a5b45f88bb",
   "metadata": {},
   "outputs": [],
   "source": [
    "# predict probabilities on the validation set\n",
    "knn_run_start = time()\n",
    "probs_knnts = knnts.predict_proba(X_valid)[:,1]\n",
    "knn_run_end = time()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "id": "48961110-164b-4d23-a908-e1bc65b9465c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Prediction run time of KNN -- 3691.000000\n"
     ]
    }
   ],
   "source": [
    "# Get the model run time\n",
    "knn_run_time = round((knn_run_end - knn_run_start))\n",
    "# evaluate scores for each threshold\n",
    "f_score_knnts = [f1_score(y_valid, to_labels(probs_knnts, t), average='weighted') for t in thresholds_cust]\n",
    "tss_score_knnts = [tss(y_valid, to_labels(probs_knnts, t)) for t in thresholds_cust]\n",
    "hss_score_knnts = [hss(y_valid, to_labels(probs_knnts, t)) for t in thresholds_cust]\n",
    "gss_score_knnts = [gss(y_valid, to_labels(probs_knnts, t)) for t in thresholds_cust]\n",
    "mcc_score_knnts = [matthews_corrcoef(y_valid, to_labels(probs_knnts, t)) for t in thresholds_cust]\n",
    "# get best threshold\n",
    "f1_ix_knnts = argmax(f_score_knnts)\n",
    "tss_ix_knnts = argmax(tss_score_knnts)\n",
    "hss_ix_knnts = argmax(hss_score_knnts)\n",
    "gss_ix_knnts = argmax(gss_score_knnts)\n",
    "mcc_ix_knnts = argmax(mcc_score_knnts)\n",
    "print('Prediction run time of KNN -- %f' % (knn_run_time))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "id": "e911652a-a481-4027-81ca-fe359aa69c6a",
   "metadata": {},
   "outputs": [],
   "source": [
    "#Choosing the best classification threshold of the KNN model\n",
    "best_knnts = thresholds_cust[hss_ix_knnts]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "id": "a2de9f97-adf9-4542-a3ef-a16ff351447d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best Threshold = 0.010\n",
      "Scores -- F1_Score: 0.931 / TSS: 0.505000 / HSS: 0.529000 / GSS: 0.360000 / MCC: 0.530015\n"
     ]
    }
   ],
   "source": [
    "# evaluate the best threshold\n",
    "f1_scorex_knnts = f1_score(y_valid, to_labels(probs_knnts, best_knnts), average='weighted')\n",
    "tss_scorex_knnts = tss(y_valid, to_labels(probs_knnts, best_knnts))\n",
    "hss_scorex_knnts = hss(y_valid, to_labels(probs_knnts, best_knnts))\n",
    "gss_scorex_knnts = gss(y_valid, to_labels(probs_knnts, best_knnts))\n",
    "mcc_scorex_knnts = matthews_corrcoef(y_valid, to_labels(probs_knnts, best_knnts))\n",
    "print('Best Threshold = %.3f' %(best_knnts))\n",
    "print('Scores -- F1_Score: %.3f / TSS: %f / HSS: %f / GSS: %f / MCC: %f' % (f1_scorex_knnts, tss_scorex_knnts, hss_scorex_knnts,\n",
    "                                                                            gss_scorex_knnts, mcc_scorex_knnts))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a8f99476-715d-4d9f-a247-123c0a0a5d40",
   "metadata": {},
   "source": [
    "### Validation set confusion matrix on the best threshold"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "id": "488eee0b-199b-4ffb-bb0a-4cf05dc54ddc",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "y_pred_class_knnts = to_labels(probs_knnts, best_knnts)\n",
    "color = 'white'\n",
    "knnts_matrix = ConfusionMatrixDisplay.from_predictions(y_valid, y_pred_class_knnts,\n",
    "                                                     cmap=plt.cm.Blues, colorbar=False)\n",
    "knnts_matrix.ax_.set(title='One Nearest Neighbor')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "60802966-bf3f-402d-9f7a-d2bd39f20d83",
   "metadata": {},
   "source": [
    "## Test set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "id": "a12970e7-d030-41d8-9cf1-abea15e7c40d",
   "metadata": {},
   "outputs": [],
   "source": [
    "# predict probabilities on the test set\n",
    "probX_knnts = knnts.predict_proba(X_test)[:,1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "id": "3f4a91eb-7188-42bb-889e-40b1d657790d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test scores -- F1_Score: 0.946 / TSS: 0.632 / HSS: 0.668 / GSS: 0.501 / MCC: 0.669\n"
     ]
    }
   ],
   "source": [
    "# evaluate on the best threshold\n",
    "f1_test_knnts = (f1_score(y_test, to_labels(probX_knnts, best_knnts), average='weighted'))\n",
    "tss_test_knnts = tss(y_test, to_labels(probX_knnts, best_knnts))\n",
    "hss_test_knnts = hss(y_test, to_labels(probX_knnts, best_knnts))\n",
    "gss_test_knnts = gss(y_test, to_labels(probX_knnts, best_knnts))\n",
    "mcc_test_knnts = matthews_corrcoef(y_test, to_labels(probX_knnts, best_knnts))\n",
    "print('Test scores -- F1_Score: %.3f / TSS: %.3f / HSS: %.3f / GSS: %.3f / MCC: %.3f' % (f1_test_knnts, tss_test_knnts, \n",
    "                                                                          hss_test_knnts, gss_test_knnts,\n",
    "                                                                          mcc_test_knnts))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "def5c750-0a19-44b6-ab09-86b7cc1c7395",
   "metadata": {},
   "source": [
    "### Test set confusion matrix on the best threshold"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "id": "d4e33049-4f96-4ed1-9bbe-68b87f6a0fa9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "color = 'white'\n",
    "y_pred_test_knnts = to_labels(probX_knnts, best_knnts)\n",
    "knnts_matrix = ConfusionMatrixDisplay.from_predictions(y_test, y_pred_test_knnts,\n",
    "                                                     cmap=plt.cm.Blues, colorbar=False)\n",
    "knnts_matrix.ax_.set(title='One Nearest Neighbor')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "55160284-feac-4009-9d8b-2b587e9bb9e8",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.11.0"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {},
   "toc_section_display": true,
   "toc_window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}
